Modelling organic carbon turnover in salt-affected soils

A thesis submitted to The University of Adelaide in fulfilment of the requirements for the degree of Doctor of Philosophy

Soils

School of Agriculture, Food and Wine

The University of Adelaide

March 2011

Raj Setia

ii

Dedicated to my parents and wife

iii

TABLE OF CONTENTS

ACKNOWLEDGEMENTS V

ABSTRACT VII

DECLARATION XI

LIST OF PUBLICATIONS XII

CHAPTER 1: REVIEW OF LITERATURE

CHAPTER 2 Manuscript 1 :

Severity of salinity accurately detected and classified on a paddock scale with high resolution multispectral satellite imagery

42

CHAPTER 3 Manuscript 2 : Is CO2 evolution in saline soils affected by an osmotic effect and calcium carbonate?

54

1.1 Introduction 1 1.2 Properties of salt-affected soils 3 1.3 Effect of salinity and sodicity on plant growth 6 1.4 Mapping of salt-affected soils using remote

sensing techniques 7

1.5 Organic matter decomposition 12 1.6 Soil organic carbon pools 14 1.7 Influence of salinity and sodicity on soil organic

carbon turnover 15

1.8 Modelling of soil organic carbon 17 1.9 Classification of soil organic carbon models 18

1.10 Evaluation, comparison and application of SOC models

20

1.11 Rothamsted carbon model 21 1.12 Modifications of RothC for subsoils, waterlogged

soils and Andosols 23

1.13 Measurable and modelled pools in RothC 23 1.14 Integration of RothC with spatial data in the

geographical information system 24

1.15 Conclusions and knowledge gaps 25 1.16 Structure of thesis 26

References 29

iv

CHAPTER 4 Manuscript 3 : Salinity effects on carbon mineralization in soils of varying texture

68

CHAPTER 5 Manuscript 4 : Relationships between carbon dioxide emission and soil properties in salt-affected landscapes

79

CHAPTER 6 Manuscript 5: Introducing a decomposition rate modifier in the Rothamsted carbon model to predict soil organic carbon stocks in saline soils

89

CHAPTER 7 Manuscript 6 : Simulation of salinity effects on soil organic carbon: past, present and future carbon stocks

99

CHAPTER 8 CONCLUSIONS AND FUTURE RESEARCH

128

APPENDIX-I 133

v

ACKNOWLEDGEMENTS

_______________________________________________________________________

I would never reach this successful accomplishment of the project without God’s grace

and help. Thanks to my parents, the engineers of my wisdom who have sacrificed their

today for my better future. My wife Deepika is my pillar of support, my source of

strength and inspiration. I can not express the gratitude I feel towards your

unwavering support.

I wish to express my gratitude to A/Prof. Petra Marschner, my principal supervisor, for

encouragement throughout the project which helped me developing a professional

approach towards scientific research and becoming a good scientist. Petra, you are an

encouraging supervisor and a nice person too; I really appreciate you for your keen

interest, unceasing encouragement and discussions which helped me to solve many of

the problems encountered during the course of my research.

I am also thankful to my other supervisors Dr. Jeff Baldock, Prof. David Chittleborough

and A/Prof. Megan Lewis. To Jeff, for valuable suggestions and discussions; to David

for his efficient communication and extended help; to Megan for guidance and support

for the remote sensing study.

I express my sincere appreciation for the contribution and help of my informal

supervisors Prof. Pete Smith and Dr. Jo Smith. To Pete for hosting me at University of

Aberdeen and helping me in carbon modelling; to Jo for introducing me the exciting

world of modelling, accepting my millions of emails and keeping me positive. Jo, your

contribution to this study is immense and without that, it would not have culminated

in this thesis.

I would like to thank Ms. Pia Gottschalk for hosting my visit to Berlin. The help and

time you gave me for the GIS run of India and Australia is greatly appreciated. You

taught me the unwritten details of RothC.

vi

I would also like to thank Drs P Rengasamy and Rob Murray for useful discussions

about salinity and sodicity which helped me in designing the experiments. This

research project would not have been possible without the support of Dr. P. K. Sharma

who granted me leave for a PhD at The University of Adelaide. He provided the

necessary facilities for collection of soil samples from the Indian site. I am thankful to

Dr. V. K. Verma for help during field work in India.

I am also thankful to Mr Sean Forrester for MIR analyses and predictions, Ms Athina

Massis for technical support, Mr Colin Rivers for help in the lab and field, Dr David

Summers, Mr Ramesh Raja Segaran and Mr Beng Umali for mapping, Dr Karen

Baumann for help in the lab, Mr Sudhir Yadav for technical help and Ms Suman Verma

for proof reading.

I am extremely grateful for the following awards: EIPRS scholarship, The University of

Adelaide Scholarship, Research abroad scholarship, Francis and Evelyn Clark Soil

Biology Scholarship, and the CRC-FFI grant which funded my visit to University of

Aberdeen and Free University, Berlin.

I would like to acknowledge and thank to Northern and Yorke Natural Resources

Management Board, CRC-FFI and Department of Climate Change for funding part of

my project.

I lovingly acknowledge my sisters and their kids and husbands for their unconditional

support. I would also like to thank Vaneet, Sumit, Poonam, Rishi and Neenu who made

me laugh and think of other things than my work.

Last but not the least, thanks are due to one and all those who happily helped me. All

these thanks are, however, only a fraction of what is due to the Almighty who blessed

me to express these words.

vii

ABSTRACT

Salinity and sodicity are major constraints for crop production in arid and semi-arid

regions of the world. Salt-affected soils cover 6.5% of the total land area of the world.

Since the global soil carbon (C) pool is greater than the atmospheric and biotic pool

combined, changes in soil organic matter content will affect atmospheric carbon

dioxide (CO2) concentration. Therefore it is important to understand soil organic

carbon (SOC) dynamics. Soil organic carbon models, which have been successfully

validated for non-saline soils, are important for estimation of past and future SOC

contents and for evaluating management effects on SOC. However, it was unclear if

they accurately predict CO2 emission/SOC stocks in salt-affected soils. In this work, an

integrated approach using remote sensing, incubation experiments, modelling and

geographical information system was used to simulate SOC dynamics in salt-affected

soils at field and regional scale in the past, present and the future.

Satellite imagery was used to map soil salinity and select soil sampling sites in

two climatically distinct regions which also differ in cause of salinity: Kadina, South

Australia and Muktsar district (Punjab), India. High resolution multispectral satellite

imagery (Quick bird, spatial resolution 0.6 m) was used to map salinity (~1:10000 scale)

in an agricultural area around Kadina, South Australia where salinity associated with

ground water or an impermeable subsoil is wide-spread. Resourcesat-I (spatial

resolution 23.5 m) was used for mapping salinity on a 1:50000 scale in Muktsar

(Punjab), India where salinity is induced by irrigation. Unsupervised classification of the

Quick bird imagery (September, 2008) covering the study area in South Australia

(hereafter called Australia) allowed differentiation of severity levels of salt-affected

soils, but these levels did not match those based on electrical conductivity (EC) and

sodium adsorption ratio (SAR) measurements of the soil samples, primarily because

the expression of salinity was strongly influenced by paddock-level variations in crop

type, growth and prior land management. Segmentation of the whole image into 450

paddocks and unsupervised classification using a paddock-by-paddock-approach

resulted in a more accurate discrimination of salinity with image derived salinity

classes correlated with EC but not with SAR. For the Indian site (hereafter called India),

viii

Resourcesat-I LISS-III data of April 2005, October 2005 and February 2006 was visually

interpreted for variation in spectral properties. The map of salt-affected soils was

generated after integration of ground and laboratory data with delineated land use

units from the satellite data. On the basis of land use and soil types, 120 (59 salt-

affected and 61 non-salt-affected) and 160 (70-salt-affected and 90-non-salt-affected)

soils were collected from 0-0.30 m depth from the Indian and Australian sites,

respectively.

Salt-affected soils occur in dry climates and often contain calcium carbonate

(CaCO3) particularly at pH > 7.5. Therefore, using CO2 emission as a measure of

microbial activity and SOC decomposition in these soils is problematic, but an

experiment involving addition of 2% wheat residues and varying the rate of calcium

carbonate added to a non-calcareous soil showed that CO2 emission from salt-affected

soils was not affected by CaCO3 addition in the presence of residues.

It has been suggested that the salt concentration in the soil solution (osmotic

potential) is a better parameter than the EC of a soil suspension to estimate the

salinity effect on plant growth. Therefore, an incubation experiment with four soils

differing in texture and amended with sodium chloride (NaCl) was conducted to assess

the effect of soil texture and osmotic potential (Os, calculated from EC and water

content) on CO2 release. The results of this study showed that, compared to saline soils

from the field, the decrease in CO2 release was greater in these soils suggesting that

the sudden increase in salinity leads to overestimation of the salinity effect compared

to saline soils in the field where salinity increases gradually. The relative decrease in

respiration was less when plotted against Os than if plotted against EC.

To investigate the importance of salinity compared to other soil properties in

soils from a salt-affected landscape, CO2 emission from the soils of India and Australia

with a wide range of EC and SAR with 2% (w/w) mature wheat residue was measured

over 120 days at constant temperature and soil water content. Cumulative CO2

emission from unamended and amended soils was related to soil properties by

stepwise regression models. Carbon dioxide release in salt-affected landscapes is

affected by EC, C availability (size of C pools) and clay content. Electrical conductivity

had a negative impact on CO2 release in soils of India and Australia, which shows the

universal effect of salinity on CO2 release, irrespective of climate and origin of salinity.

ix

Therefore, there is a need to add a decomposition rate modifier for salinity in the SOC

models for accurate prediction of SOC dynamics and CO2 release from salt-affected

soils.

The Rothamsted Carbon Model (RothC) was modified to take into account the

reduced plant inputs into salt-affected soils. Plant inputs were calculated based on a

generalised equation from the literature. The decomposition rate modifier for salt-

affected soils was based on the comparison of measured and modelled CO2 emissions

from wheat residue amended soils of India and Australia. The modelled CO2 emissions

were higher than measured CO2 emissions. In order to match the measured and

modelled CO2 emissions, rate modifiers ranging from 0.2-1 were introduced in the

model. After accounting for the laboratory effect due to soil disturbance, the impact of

salinity (calculated using Os) or sodicity (measured as SAR) on the rate of

decomposition was calculated. A significant positive relationship was found between

decomposition rate modifier and Os whereas SAR had no effect. Therefore, a

decomposition rate modifier due to salinity (as a function of Os) was introduced into

RothC.

The RothC with the plant input modifier and decomposition rate modifier was

used to estimate past SOC content when saline soils were non-saline and future SOC

content. These simulations were performed for the Indian and Australian sites. The

results showed that the modelled past SOC when the soils were non-saline was higher

than measured SOC of saline soils; thus these soils have lost SOC (31 t ha-1 for India

and 55 t ha-1 for Australia). On the other hand, simulations with the decomposition

rate modifier only, without taking into account the reduced plant input, suggest that

SOC of saline soils has increased since they became saline. Since SOC in saline soils is

lower than in non-saline soils, this shows that in order to accurately model SOC stocks

in saline soils, both reduced plant inputs and reduced decomposition rate have to be

taken into account. Overall SOC content was more strongly affected by reduced plant

inputs than by reduced decomposition rates. In addition, future SOC stocks of India

and Australia were simulated with and without modifiers from 2009-2100. In saline

soils of both regions, the simulation of SOC without modifiers showed that, compared

to the present SOC content, SOC would decrease by ≤15% by the year 2100, whereas

x

simulations with decomposition rate modifier and plant input modifier indicate that

SOC would decrease by 39% for the Indian site and by 29% for the Australian site.

The key findings from the research are:

I. High resolution multispectral imagery with paddock-by-paddock approach

allowed accurate mapping of different levels of salinity severity.

II. In saline soils, osmotic potential is a better measure to assess the impact of salt

on microbial activity than EC, particularly when comparing soils of different

texture.

III. In soils from salt-affected landscapes, salinity and reduced carbon availability

determine CO2 emission.

IV. Two novel approaches were developed: (a) calculation of a decomposition rate

modifier from incubation experiments after taking into account the laboratory

effect and (b) calculation of past SOC content when saline soils were non-

saline.

V. The predictions of SOC stocks from saline soils have been overestimated by not

taking into account the negative effect of salt on decomposition rate and plant

inputs.

VI. For realistic modelling of SOC stocks and turnover in saline soils, both reduced

decomposition rate and reduced plant inputs need to be considered.

xi

DECLARATION

This work contains no material which has been accepted for the award of any other

degree or diploma in any university or other tertiary institution and, to the best of my

knowledge and belief, contains no material previously published or written by another

person, except where due reference has been made in the text.

I give consent to this copy of my thesis when deposited in the University Library, being

made available for loan and photocopying, subject to the provisions of the Copyright

Act 1968.

The author acknowledges that copyright of published works contained within this

thesis (as listed below) resides with the copyright holder(s) of those works.

I also give permission for the digital version of my thesis to be made available on the

web, via the university’s digital research repository, the Library catalogue, the

Australasian Digital Thesis Program (ADTP) and also through web search engines,

unless permission has been granted by the University to restrict access for a period of

time.

Raj Setia Date :

xii

LIST OF PUBLICATIONS

Setia, R., Marschner, P., Smith, P., Baldock, J., Chittleborough, D., Smith, J., 2010. Using salt-amended soils to calculate a rate modifier for salinity in soil carbon models, 19th World Congress of Soil Science, Soil solutions for a changing world, Brisbane, Australia. Setia, R., Lewis, M., Marschner, P., Raja Segaran, R., Summers, D., Chittleborough, D., 2011. Severity of salinity accurately detected and classified on a paddock scale with high resolution multispectral satellite imagery. Land Degradation & Development, DOI: 10.1002/ldr.1134. Setia, R., Marschner, P., Baldock, J., Chittleborough, D., 2010. Is CO2 evolution in saline soils affected by an osmotic effect and calcium carbonate? Biology and Fertility of Soils 46, 781-792. Setia, R., Marschner, P., Baldock, J., Chittleborough, D., Smith, P., Smith, J., 2011. Salinity effects on carbon mineralization in soils of varying texture. Soil Biology & Biochemistry 43, 1908-1916. Setia, R., Marschner, P., Baldock, J., Chittleborough, D., Verma, V., 2011. Relationships between carbon dioxide emission and soil properties in salt-affected landscapes. Soil Biology & Biochemistry 43, 667-674. Setia, R., Smith, P., Marschner, P., Baldock, J., Chittleborough, D., Smith, J., 2011. Introducing a decomposition rate modifier in the Rothamsted carbon model to predict soil organic carbon stocks in saline soils. Environmental Science & Technology, dx.doi.org/10.1021/es200515d. Setia, R., Smith, P., Marschner, P., Gottschalk, P., Baldock, J., Verma, V., Smith, J., 2011. Simulation of salinity effects on soil carbon: past, present and future carbon stocks. Agriculture, Ecosystems and Environment, Submitted.

CHAPTER 1

INTRODUCTION AND REVIEW OF LITERATURE

1.1 Introduction

Approximately 6.5 % of the world land area is affected by either salinity or sodicity

(Szabolcs, 1993) and Pessarakli and Szabolcs (1999) reported that salt-affected land

comprised 19 percent of the 20.8 billion hectares of arable land on earth. Saline and

sodic soils account for approximately 40 and 60 percent of world’s salt-affected land,

respectively (Tanji, 1990).

Salt accumulation in soils is a major threat to agricultural production and

ecosystem sustainability because it reduces plant growth and increases the risk of

erosion. On a global scale, the cost to agriculture productivity from salinity is estimated

at $US 12 billion a year and this cost is expected to increase with further increases in

the area affected by salt in the future (Ghassemi et al., 1995).

One important process affected by salinity and sodicity is soil organic matter (SOM)

turnover which is mainly mediated by microbial activity. Soil microorganisms carry out

crucial functions in all soils; most importantly, they are the key drivers of soil organic

matter decomposition and nutrient cycling. Stresses such as salinity will lead to the

inhibition or death of sensitive microorganisms and, may irreversibly, reduce essential

microbial activities and thereby affect SOM decomposition. Soil organic matter

turnover plays a key role in greenhouse gas emissions, soil structural stability, ion

exchange, water quality and ecosystem sustainability. In general, microbial activity and

plant inputs are lower in salt-affected soils (Rietz and Haynes, 2003). If inputs of soil

organic carbon (SOC) in salt-affected soils were similar to those in non-saline soils,

reduced decomposition rates may help in mitigating the green house effect by

increasing carbon (C) sequestration in these soils and reducing the emission of carbon

dioxide (CO2) into the atmosphere.

Reclamation of salt-affected soils would enhance soil quality, improve productivity

and SOC sequestration (Lal, 2001), but for a better understanding and quantification

of their SOC turnover and sequestration potential, integrated use of frontier

technologies such as remote sensing, modelling and sophisticated analytical methods

is required. Soil organic carbon models may help in predicting and understanding

2

future changes in SOC in response to changing climate, altered land use and different

land management practices. Soil organic carbon models have been used successfully to

predict changes in soil organic matter content in non-saline land in the short or long-

term plot and on different spatial scales: field (Jenkinson et al., 1987; Parton, 1996;

Franko et al., 1997; Li et al., 1997; Smith et al., 1997; Skjemstad et al., 2004; Gottschalk

et al., 2010), regional (Ardo and Olsson, 2003) as well as country (Smith et al., 2005;

Smith et al., 2007; Smith et al., 2010). However, these models have not been calibrated

for salt-affected soils. The combination of remote sensing and Geographical

Information System (GIS) data with process based models (Rothamsted Carbon Model,

RothC), CENTURY, (DeNitrification and DeComposition (DNDC), DAISY etc.) can provide

tools for understanding spatial C dynamics, but this type of integration has not been

used for studying SOC in salt-affected landscapes.

The following literature review will cover (1) properties of salt-affected soils

and how these soils can be mapped (2) effects of salinity and sodicity on SOC (3)

modelling of SOC and (4) integration of modelled SOC data with GIS (Figure 1).

Figure 1: Overview of topics covered in the literature review and how they relate

to soil organic carbon dynamics in salt-affected soils

Climate and ancillary data

Soil organic carbon dynamics

Remote Sensing Mapping of salt-

affected soils

Modelling

Soil organic matter

3

1.2 Properties of salt-affected soils

Salt accumulates when the loss of salt via leaching is less than the addition of salt

through rainfall or irrigation, or as a result of a rise of saline groundwater. The US

Salinity Laboratory Staff (1954) suggests the following three classes of salt-affected

soils:

(A) Saline soils: Salinity is measured in units of electrical conductivity (EC) using a

range of soil/water ratios: saturated paste of soil (ECe), 1:1, 1:2 or 1:5. The various EC

values can be converted to ECe using a multiplication factor based on soil texture

(Shaw, 1999). Saline soils have a high concentration of soluble salts and an electrical

conductivity of saturation extract (ECe) greater than 4 dS m-1 (Figure 2).

There are two forms of salinity:

(a) Primary Salinity caused by soluble salts originating from the weathering of primary

minerals, aeolian recycling or cyclic accession, where salts are transported inland by

winds off the ocean.

(b) Secondary salinity results from human activities such as irrigation and land clearing

in areas that are not irrigated (dryland salinity) which causes a change in the salt and

water balance.

In general, saline soils are flocculated but the high salt content in the soil has an

adverse affect on plant growth due to (1) low soil osmotic potential, causing water to

move from areas of lower salt concentration (plant tissue) into the soil where the salt

concentration is higher thereby inducing water stress in plants (2) specific ion toxicity,

and 3) ion imbalance (Ca2+, K+ and Na+) all of which disrupt plant metabolism.

(B) Sodic soils: Sodicity of a soil is expressed using the exchangeable sodium

percentage (ESP):

CEC

]100[Na ESP ex

in %

Where,

CEC = cation exchange capacity in cmol p (+) kg-1

[Na+] = exchangeable Na+ in cmol (+) kg-1.

4

Figure 2: Categories of salt-affected soils based on electrical conductivity (ECe) and sodium adsorption ratio (SARe) measured in saturated soil: water extract [Source: US Salinity Laboratory Staff (1954)]

For determination of ESP, CEC is required but its determination is laborious and

time consuming. Therefore, a convenient measure to express the sodicity of a soil

solution or of water is the sodium absorption ratio (SAR) which is calculated as follows:

0.5225:1

])[Mg]Ca ([

][Na SAR

Where, [Na+], [Ca2+], [Mg2+] = concentration of Na+, Ca2+ and Mg2+ in the soil

solution in mmol l-1. Note that SAR does not have units.

Sodic soils typically have a high pH (>8.5) and high (>13) SAR but ESP varies with

classification system. For example, the US system defines a sodic soil as that with an

ESP greater than 15 (Soil Survey Staff, 1998) whereas the recent Australian

classification system (Isbell, 2002) states that sodic soils have an ESP greater than 6

(Rengasamy, 2006). Sodium will replace Ca and Mg on the clay particles, which can

cause dispersion of clay and loss of soil structure. Dispersed clay can move into soil

pores, leading to blockage of pores, which, in turn, results in poor drainage and

reduced aeration (Rengasamy and Olsson, 1991) and hence, reduced plant growth. As

sodic soils dry out, they become very hard and the decrease of macrospores and pore

connectivity reduces the capacity of roots to penetrate the soil (Marschner, 1995).

NOTE: This figure is included on page 4 of the print copy of the thesis held in the University of Adelaide Library.

5

(C) Saline- sodic soils: These soils have an ECe greater than 4 dS m-1 and SAR greater

than 13 but their pH is less than 8.5. In general, these soils have good soil structure and

water movement through the soil profile is adequate. They can have the

characteristics of either a saline or a sodic soil, depending on whether sodium or

calcium dominates.

In general, soils are often first saline and then become sodic as Na replaces Ca

and Mg from the cation exchange sites which causes dispersion of clay particles and

thereby affects oxygen availability in the soils, but sodic soils that are also highly saline

contain high concentrations of electrolytes and these soils remain flocculated. This can

be explained by the three plate model of Quirk (2001). In a clay domain, clay crystals

overlap to a certain extent and slit-shaped pores form within the domain or where

crystals do not overlap. A repulsive pressure operates over the surface area of the slit-

shaped pores, while an attractive pressure operates over the surface area of the clay

crystals. A high concentration of monovalent cations such as Na+ results in the

accumulation of these dispersive cations in the slit-shaped pores and these ions do not

form extensive double layers as compared with the smaller double layers for cations of

higher valence (Sumner et al., 1998). Thus, the attractive force can override the

repulsive force in soil systems resulting in the flocculation.

In Australia, irrigation-induced salinity, dry-land salinity and dry saline land

salinity are wide spread. Dryland salinity commonly occurs at the foot of slopes and in

valley floors where the water table is shallow (Rengasamy, 2006). This salinity is

caused by agriculture by clearing of native vegetation which disturbs the water

balance. Cultivation of annual crops leads to lower evapotranspiration rates than

under native perennial plants, which results in rising of the ground water table and

dissolved salts and, hence dryland salinity. On the other hand, dry saline land or

transient salinity is not associated with the ground water table and involves surface

and sub-surface soil salinity. This type of salinity was first identified in 1942 by Herriot

and is also termed ‘magnesia patches’. Dry saline land generally occurs within duplex

or texture-contrast soils on upper or middle slopes, particularly when there is a

sandy/loamy A horizon over an impermeable, clay rich, sodic B horizon. It results from

seasonal movement of salt in the top soil due to evaporation and rainfall (Rengasamy,

2002).

6

1.3 Effect of salinity and sodicity on plant growth

Salinity inhibits plant growth and the activity of soil microorganisms as a result

of low water availability (due to low osmotic potential of the soil solution), ion toxicity

(Na, Cl) and nutrient imbalance (Marschner, 1995). Low osmotic potential (more

negative) draws water from cells into the soil solution causing cell dehydration and

also makes it harder for plants or microorganisms to extract water from soil. Further,

high concentrations of Na+ and Cl- ions in saline soils can have a toxic effect on plant

roots; excessive uptake of Cl- ion reduces photosynthesis and respiration resulting in

poor growth and possibly premature death of plants (Tester and Davenport, 2003).

Sodic soils have poor soil structure which can adversely affect root growth and plant

physiology due to poor aeration, lack of pores and, boron and bicarbonate (HCO3-)

toxicity at high pH.

Many studies have shown a decrease in yield of crops as a result of salinity

stress. For example: rice (Bajwa et al., 1986), wheat (Richard, 1983; Rengasamy, 2010),

cotton (Meloni et al., 2001), maize (Bajwa et al., 1986) and sugar beet (Ghoulam et al.,

2002). The impact of salinity on crop yield depends on the soil water content and

environmental factors. The plants growing in saline soils under hot and dry conditions

are more sensitive to salinity than when growing in a cool and humid environment due

to differential water content and thus osmotic potential. The degree to which plant

growth is affected by salinity differs between crops. Maas and Hoffman (1977)

prepared a comprehensive table listing slope, i.e. percent reduction in relative yield

per increase in ECe (dS m-1) and threshold salinity levels (critical level of salinity beyond

which the yield declines) for a wide range of crops using a piecewise linear model.

They developed the following equation for expected relative yield of a crop:

Y = 100 – B (ECe – A)

Where,

Y = relative yield in %

B = threshold value (dS m-1)

A = slope of the linear line

ECe = average root zone soil salinity (dSm-1)

The threshold and slope used to calculate relative yield for globally important

crops (as defined by Parihar et al., 2009) due to salinity are given Table 1. Although

7

barley and sunflower have lower threshold values than canola, they can be considered

to be more salt-tolerant than canola because their yield declines more slowly (small

slope). It should be noted that within each species, there can be considerable

differences in salt tolerance among cultivars.

Table 1: Relationship between yield and salinity of important crops

(Source: Maas and Hoffman, 1977)

1.4 Mapping of salt-affected soils using remote sensing techniques

To assess the extent of salinity and sodicity and the rate of change in the size of

the salt-affected area, it is important to map salt-affected soils accurately.

Traditionally, field survey methods were used for monitoring of soil conditions but the

cost and time for these surveys makes them uneconomical. Remote sensing is the

process of identification, delineation, measurement of surface features and their

processes from a distance without directly coming into physical contact. It can provide

multi-spectral, multi-sensor and multi-temporal data which can generate accurate,

timely and cost-effective information on natural resources (Liu and Kogan, 2002).

Remote sensing measures electromagnetic energy from the sun, emitted by the

earth or by the sensor itself (microwave sensor) which is reflected, scattered or

emitted by different surface features on the earth. Remote sensing makes use of the

NOTE: This table is included on page 7 of the print copy of the thesis held in the University of Adelaide Library.

8

electro-magnetic spectrum which extends from gamma rays (< 0.03 nm), through X

rays (0.03-300 nm), ultraviolet (0.30 -0.38 µm), visible (0.38-0.72 µm), infrared (0.72-

15.0 µm), microwave (1-300 µm) to radio waves (> 30 cm). Remote sensing systems

exist in the entire visible region, some wavelength bands in the infrared and the

microwave region, because electromagnetic radiation in these regions can penetrate

the atmosphere through so-called atmospheric windows. Among these regions, the

optical wavelength region (0.38-15 µm) is an important region for remote sensing

applications which is further subdivided into two regions: reflective (0.38-3.0 µm) and

far infrared (thermal or emissive, 3.0 -15.0 µm). The reflective portion is subdivided

into visible (0.38-0.72 µm), near infrared (0.72-1.3 µm) and mid infrared (1.3-3.0 µm).

Further, optical/thermal imaging systems can be classified according to the number of

spectral bands used: monospectral or panchromatic (single wavelength band, "black-

and-white", grey-scale image), multispectral (several spectral bands), superspectral

(tens of spectral bands) and hyperspectral (hundreds of spectral bands) systems.

Aerial photography, one of the most common and versatile forms of remote

sensing, is still popular for mapping of salt-affected soils due to its capability to provide

historical information. The maps generated from photo interpretation are used to

divide regions into landscape units which may contain several soil characteristics and

salinity levels. However, the location and extent of salinity within units can not be

accurately assessed with the aerial photographs because of uncertainties about the

approximate scale at which salinity is expressed in relation to the scale and theme of

the mapping. With the advent of satellite remote sensing in the early 70s after the

launching of the Landsat satellites by National Aeronautics and Space Administration

(NASA), USA; the SPOT satellite by France and the Indian remote sensing satellite (IRS)

by the Indian Space Research Organization (ISRO), there has been an increasing use of

satellite imagery for inventory and monitoring of natural resources world-wide.

Satellite-borne optical sensors record the intensity of electromagnetic radiation

reflected from the earth at different wavelengths. The sensors are characterized by the

bands at which they measure the reflected energy. Each object has its own unique

'spectrum', some of which are shown in the Figure 3.

9

Figure 3: Typical spectral reflectance curves for vegetation, soil and water

(Source: www.microimages.com/documentation/Tutorials/hyprspec.pdf)

Remote sensing relies on the fact that particular features of the landscape such

as crops, salt-affected land and water reflect light differently in different wavelengths

(Figure 3). For example, vegetation appears green because it reflects green light more

than other visible wavelengths which are strongly absorbed. This can be seen as a peak

in the green band in the reflectance spectrum for vegetation in Figure 3. The spectrum

also shows that vegetation reflects even more strongly in the near infrared part of the

spectrum because the chlorophyll in a plant absorbs visible and especially red light for

photosynthesis, whereas near infrared light is reflected very effectively because plants

do not use it for energy generation.

Monitoring and mapping of salt-affected soils can be achieved by satellite

remote sensing, field and laboratory measurements, and the use of GIS for processing,

transforming and displaying the data (Metternicht and Zinck, 2009). Salt on the soil

surface in the form of crusts or efflorescence is not readily sensed because it does not

have distinctive spectral features and is generally mixed with other soil minerals.

Instead, the symptoms of the salinity are sensed: reduced plant vigour, reduced cover

NOTE: This figure is included on page 9 of the print copy of the thesis held in the University of Adelaide Library.

10

and growth, change in vegetation composition, greater soil exposure and possibly

brighter soil surfaces (Lewis, 2002, Dutkiewicz et al., 2006).

For mapping of salt-affected soils, the most important task is to select the

appropriate time for collecting the satellite imagery. Venkataratnam (1983) mapped

soil salinity in Punjab, India, at different times of the year (pre-monsoon, post-

monsoon and harvest seasons) using Landsat-MSS images and concluded that the

spectral curves of highly and moderately saline soils change considerably throughout

the year, which complicates the selection of the appropriate time for acquiring satellite

imagery for accurately mapping of salt-affected land. Johnston and Barson (1990)

found that discrimination of saline areas in Australia was most successful during peak

vegetation growth due to low fractional vegetation cover in saline areas which,

however, may not be distinguishable from poor plant cover caused by grazing, erosion

or nutrient deficiency. Similarly, Verma et al. (1994) found the satellite data acquired

during summer (March and first week of April) at crop maturity was significantly better

for mapping of salt-affected soils in India due to maximum biomass which helps

distinguishing between salt-affected and non- salt-affected land. On the other hand,

Siderius (1991) found that salinity was most clearly expressed at the end of the

irrigation or rainy season when plots were bare. Kalra and Joshi (1996) evaluated

Landsat (MSS and TM), SPOT and IRS (LISS-I and II) images acquired during summer

(April, May), winter (January /February) and autumn (October) for delineation of salt-

affected soils in Rajasthan, India and concluded that moderately and severely salt-

affected soils could be mapped in all seasons using false colour composite (FCC).

However, the summer season images provided the best discrimination of salt-affected

soils.

Over the last two decades, several studies have assessed the use of

multispectral and hyperspectral imagery for mapping and monitoring of salt-affected

soils (Saha et al., 1990; Verma et al., 1994; Dwivedi, 1996; Metternicht and Zinck, 1997;

Metternicht and and Zinck, 1998, Evans and Caccetta, 2000, Metternicht and Zinck,

2003, Dutkiewicz et al., 2009). Various approaches such as false-colour composites and

visual interpretations (Rao et al., 1991; Joshi and Sahai, 1993), principal component

analysis (Dwivedi, 1996), digital classification (Steven et al. 1992; Vincent et al., 1996)

11

and greenness, brightness and classification of salinity symptoms in trees (Vidal et al.,

1996) have been used for mapping soil salinity. In general, bands in the near and mid

infrared region provide the best information about soil moisture and salinity

(Metternicht and Zinck, 2003). Multispectral imagery is not useful for mapping salinity

if saline soils are covered with halophytic vegetation (Howari, 2003). Moreover, bright

or white patches can indicate exposed non-saline soils and not necessarily salt

efflorescence (Metternicht and Zinck, 2009). Studies have attempted to combine soil

and terrain data to overcome this problem (Metternicht and Zinck, 2003, Dutkiewicz et

al., 2009).

Depending on the scale of expression of salinity symptoms, spatial resolution

has a significant effect on the capacity to identify salt-affected soils and crops.

Manchanda (1984) observed that Landsat-MSS was of limited use for identification of

saline areas for site specific management due to its low spatial resolution, but Dwivedi

(1996) used principal components analysis of Landsat-MSS data for monitoring salt-

affected soils of the Indo-Gangetic alluvial plains. Joshi and Sahai (1993) compared the

accuracy of TM, MSS, and SPOT and found that Landsat TM was superior for soil

salinity mapping at regional level. Dwivedi et al. (2009) found the IKONOS-II

multispectral data with 4 m resolution was better than IRS-1D LISS-III (spatial

resolution 23.5 m) and pan sharpened LISS-III (spatial resolution 5 m) for mapping of

saline-sodic soils at field level.

The spectral resolution of satellite data is important for the differentiation of

different categories of salt-affected soils. Metternicht and Zinck (1997) combined the

Landsat TM optical bands 1, 2, 4, 5, 6 and 7 to obtain maximum separability between

saline and sodic soils using digital classification, field observations and laboratory

measurements. Dutkiewicz et al. (2009) evaluated the performance of Hyperion

imagery (spatial resolution 30 m and 242 bands covering the 400-2500 nm spectral

range) for mapping surface symptoms of dryland salinity in southern Australia and

found that the hyperspectral imagery was unable to differentiate halophytic samphire

vegetation (an indicator of salinity) at slight or moderate levels of salinity but could be

used to map high to very high and extremely high salinity.

Although multi- and hyperspectral images are useful tools for mapping the surface

expression of salinity, soil texture and moisture as well as organic matter content

12

interfere with spectral signatures of salinity (Metternicht and Zinck, 1997). Besides

this, the colour and brightness of soil surfaces have a strong impact on the spectral

signature. For example, after rainfall, the optical properties of soils with different clay

content vary as they retain water to a different extent which results in confusion of

spectral signatures between low or medium salt-affected soils and non-affected soils

(Goldshleger et al., 2001, Bannari et al., 2008).

Thus, remote sensing can be used effectively to map salt-affected soils.

Multispectral images with low spatial resolution can be used to map salinity in larger

areas which helps policy makers and land managers to obtain information about the

extent of salinity. Hyperspectral imagery is generally used in smaller areas for

delineating the salinity, but the higher cost to acquire the data and the more complex

image calibration limits the wide-spread use of this imagery.

1.5 Organic matter decomposition

Globally soils constitute the third largest global C pool after oceanic (38 000 Pg)

and geologic (5000 Pg) pools (Lal, 2004). The SOC pool, estimated at 1550 Pg to 1

meter depth is about twice the atmospheric pool (760 Pg) or 2.8 times the biotic pool

(560 Pg) (Lal, 2008).

Organic matter plays an important role in improving physical, chemical and

biological characteristics of soils. For example, organic matter improves soil structure

by binding to clay particles via cation bridges and through stimulation of microbial

activity and root growth. Organic matter is a source of plant nutrients and can retain

nutrients in a plant-available form. Organic matter indirectly improves soil structure by

increasing microbial activity and root growth as microbial slimes, fungal hyphae and

roots bind aggregates together (Tisdall and Oades, 1982). Addition of organic matter

can improve soil structure in sodic soils but marked improvement of soil structural

stability is often only observed when Ca (e.g. in the form of gypsum) is added together

with organic matter (Nelson et al., 1999) because Ca provides the cation bridges that

bind clays and organic matter particles both of which are negatively charged.

Baldock and Nelson (1999) defined SOM as the sum of the living and dead organic

material found in a soil or on its surface, regardless of its origin or state of

decomposition but excluding the above ground portion of living plants. Soil organic

13

matter consists of C, hydrogen (H), oxygen (O), nitrogen (N), phosphorus (P) and

sulphur (S). It is difficult to measure SOM, there are several reliable methods to

determine SOC which can then be converted to SOM by the following formula

SOM (%) = SOC (%) x 1.724

Soil organic carbon can be divided into two groups, living (5% of SOC) and non-

living (95% of SOC) components (Theng et al., 1989). The living components include

plant roots, macro organisms and micro organisms whereas the non-living components

include free macro organic matter, protected organic matter, humus and char.

Carbon sequestration in soil, which occurs when C inputs are greater than losses,

can be an important strategy for reducing atmospheric CO2 concentrations. The inputs

of organic matter include manures and plant residues. The loss of SOC from the soil

includes respiration that occurs during decomposition of organic matter, leaching of

organic carbon into the ground water, erosion of organic matter-rich top soil and

burning of crop residues. Soil organic matter decomposition is a complex multistage

process which is mediated by microorganisms. More than 90 % of SOM is decomposed

by microorganisms (bacteria and fungi), but protozoa, earthworms, nematodes,

molluscs and arthropods also play also a significant role. They create an environment

conducive for microbial decomposition of organic matter. In non salt-affected soils, it

has been shown that bacteria usually dominate the initial phases of decomposition

because they can grow rapidly on easily available compounds (e.g. soluble sugars,

carboxylates) that are present in fresh plant residues. Because of

mineralization/utilization by bacteria, the concentration of these easily available

substrates decreases over time until only more recalcitrant material (e.g. cellulose

encrusted with lignin) remains. Fungi, which grow more slowly, but are able to

decompose more recalcitrant material that is largely unavailable to bacteria, dominate

in the later phases of the decomposition process (Cahyani et al., 2002 ; Rantalainen et

al., 2004). The turnover time of these compounds varies from 14 days for

hemicellulose to 500 days for lignin (Killham, 1994). Complete decomposition of

organic material is rarely achieved as some of the organic material is humified and

becomes protected by binding to clays or occlusion in soil aggregates which increases

turnover rates to thousands of years.

14

1.6 Soil organic carbon pools

Although SOC consists of many different compounds, it can be conceptually

divided into a number of pools which differ in magnitude and turnover times. The

turnover times represent the times carbon resides in a certain pool and hence is a

measure of the stability of the pools (Jastrow et al., 2007). These pools are conceptual

(Jenkinson and Rayner, 1977), and different authors use different terminology, but it is

recognized that they play different functional roles in the soils. The identification of

different pools is important in terms of stabilization mechanisms (physical and

chemical and bioavailability) and turnover as well as understanding long- and short-

term changes in SOC.

Commonly, these pools consist of an active fraction, a slow fraction and a

resistant fraction. The active (labile) or fast pool is composed of fresh plant and animal

residue and easily decomposable compounds from leaf litter and roots with short

turnover times (from weeks to years). The active fraction that is not decomposed

moves into the slow or resistant fractions. The slow pool, partially resistant to

decomposition, consists of cells and tissues of decomposed material with turnover

times from 10 to more than 100 years, whereas the resistant or inert or passive pool

has a turnover time of 1000 years or more and is composed of highly humified organic

compounds (Trumbore et al., 1989). Typically the size of the various pools is <10% for

the active fraction, 40-80% for the slow fraction and 10-50% for the passive fraction

(Amundson, 2001).

Baldock and Skjemstad (1999) divided non living SOM into four pools: dissolved

organic matter (DOM), particulate organic matter (POM), humus and inert organic

matter (IOM). Dissolved organic matter is usually defined as the organic matter in soil

solution with < 0.45 µm diameter whereas POM includes organic matter >53 µm with a

recognisable structure (macro organic matter). Humus consists of non-humic (30% of

humus) and humic substances. Non-humic substances include carbohydrates, lipids,

organic acids, pigments and proteins, whereas humic substances have complex

chemical structure and can be divided into three fractions namely fulvic acid (soluble in

acid and alkali), humic acid (soluble in alkali) and humin (not soluble in either medium).

The inert organic matter includes highly carbonaceous materials such as charcoal,

graphite and coal. In Australian soils, inert SOM is primarily charcoal (Skjemstad et al.,

15

1998). Particulate organic matter is often considered the active pool (Baldock and

Skjemstad, 1999), but it is also considered as an intermediate SOM fraction falling

within both active and slow pools (Brady and Weil, 1999). The turnover time of these

measurable pools is <10 years for POM, tens of years for humus and hundreds to

thousands of years for IOM.

1.7 Influence of salinity and sodicity on soil organic carbon turnover

Salinity and sodicity decrease plant productivity and consequently reduce organic

matter input. But they also affect microbial activity and therefore can change organic

matter decomposition rates. The main problem encountered by microorganisms in

saline soils is the low osmotic potential of soil water outside of the cells, which causes

cells to lose water. Salinity-resistant microorganisms which can be found in Archaea,

Bacteria and Eucarya (Oren, 1999); they can rapidly accumulate salts such as KCl or

compatible solutes to adjust their intracellular osmotic potential (Oren, 1999). Salt

resistance mechanisms are energy-intensive and may therefore affect the efficiency of

C utilization by microorganisms. Oren (1999) calculated that the synthesis of organic

solutes and Na extrusion from the cells would require up to three times more energy

than cell wall synthesis. Indeed, C utilization efficiency (conversion of substrate C into

microbial biomass) decreases with increasing salinity (Rietz and Haynes, 2003),

resulting in greater CO2 release per g substrate C. It has been shown that salinity

changes microbial community structure (Pankhurst et al., 2001; Gros et al., 2003) and

that fungi were in general more sensitive to salinity than bacteria (Pankhurst et al.,

2001; Gros et al., 2003), thus the bacteria/fungi ratio was higher in saline soils. This

may have important implications for organic matter decomposition because fungi play

a key role in decomposition of polymeric plant and animal residues as well as

recalcitrant compounds such as lignin (Killham, 1994).

Both salinity and sodicity may alter physico-chemical properties of soils which

affect transformation of added plant and native C in soil. Salinity decreased SOC

mineralization (Laura, 1974, 1977; McCormick and Wolf, 1980; Sarig and Steinberger,

1994; Garcia and Hernandez, 1996; Pathak and Rao, 1998; Sardinha et al., 2003;

Wichern et al., 2006; Tripathi et al., 2006; Ghollarata and Raiesi, 2007; Yuan et al.,

2007; Walpola and Arunakumara, 2010). On the other hand, there are contradictory

16

reports on the effect of sodicity on SOC decomposition. Sodicity can increase

mineralization of added C due to dispersion of soil which releases organic matter from

the inside of aggregates which then becomes accessible to microorganisms (Nelson et

al., 1996). Contrarily, a negative effect of sodicity on C mineralization has also been

reported which may be due to the quality of the organic matter and anaerobic

conditions in sodic soils (Abdou et al., 1975). Kaur et al. (2002) found a decrease in

microbial biomass C in soils receiving continuous sodic irrigation water for 19 years in

rice-wheat cropping system. Rietz and Haynes (2003) reported a stronger detrimental

effect of salinity compared to sodicity on soil biological properties. They found an

exponential negative relationship between microbial biomass C and EC but a linear

negative relationship of microbial biomass C with ESP and SAR. In salt-affected soils, C

and N mineralization was inhibited at higher EC with nitrification being more sensitive

than ammonification (McClung and Frankenberger, 1987; Rasul et al., 2006).

Dendooven et al. (2010) showed that decomposition of easily decomposable organic

material (such as glucose) was inhibited in an extremely alkaline saline soil.

In saline-sodic soils, decomposition is affected by both salinity and sodicity.

Wong et al. (2008) found the highest respiration in low salinity-high sodicity soils and

the lowest in mid salinity-low sodicity soils. Further, Wong et al. (2009) found that soil

microbial biomass and respiration increased initially after addition of organic material

to a highly saline sodic soil and they explained this contrasting effect by the increased

ionic strength of solution which causes release of SOC sorbed onto aggregates. Skene

and Oades (1995) found that treating a soil with a solution with low SAR and high EC

decreased the concentration of soluble carbon.

Salt-affected soils contain various salts but sulphate and chloride salts of

sodium dominate. Chloride and sulphate salts may have a differential effect on C and N

mineralization in soils (Agarwal et al., 1971; Broadbent and Nakashima, 1971). Chandra

et al. (2002) found increased C mineralization with increasing KCl and K2SO4 salinity. Li

et al. (2006 a) reported that cumulative CO2 evolution decreased with increasing NaCl

concentration, but increased with Na2SO4 salinity. Further, Li et al. (2006 b) reported

that the effect of salinity on C mineralization depended on the duration of incubation

period : CO2 evolution decreased with increasing NaCl in the first 1-3 days of

17

incubation but the effect was variable thereafter, whereas Na2SO4 had no effect on

CO2 evolution during the first 1-3 days and had a variable effect thereafter.

Most of these studies were carried out in salt-amended soils which may not be

a true representation of ionic composition of naturally occurring salt-affected soils.

Moreover, addition of salts to a non-saline soil may not allow microorganisms to adapt

to salinity as they would in the field where salinity develops more gradually. McClung

and Frankenberger (1987) found that C mineralization was similar in naturally saline

soils with ECe 44 dS m-1 and salt-amended soils with a ECe 10, 15 and 20 dS m-1. This

indicates that using salt-amended soils may lead to overestimation of the salinity

effect on SOM decomposition.

1.8 Modelling of soil organic carbon

A model is a mathematical representation of the real system with well defined

boundaries. Turchin (1998) classified models into three types according to the goal of

the model in question.

1) Descriptive models –are used to identify and obtain information on the

characteristics of a particular issue and merely describe something mathematically.

Common statistical models in this category include the mean, median, mode,

range, and standard deviation.

2) Explanatory models –are used to understand phenomena by describing

relationships among variables in a system

3) Predictive models –are used to predict a certain phenomena on the basis of

hypotheses and general relationships among variables.

Among these models, the predictive models are most valuable for SOC, but also

the most challenging.

The use of SOC models is an important research tool to understand SOC dynamics

and develop management strategies to optimize inputs for C sequestration. Models

can

1. quantify expected results

2. help understand the contribution of different underlying mechanisms to the

observed results

3. explore the relationship between various components simultaneously

18

4. make future projections of changes in SOC under different management and/or

climate scenarios

5. extrapolate results to other situations within the calibration conditions

1.9 Classification of soil organic carbon models

Models can be classified based on input, output, scope and application.

Different classification schemes have been proposed to compare different SOC models.

Jenkinson (1990) classified SOC models into the following categories:

(1) Single homogenous compartment models: These models assume a single

compartment and have been applied mainly to organic N and C. Mathematically, they

can be written as: C = Co (1-e-kt) where C is total C mineralized over a given time period,

Co is the potentially mineralizable fraction of organic C, k is the proportional rate

constant. This equation assumes that of the amount of organic material in the soil

declines exponentially with time and the relative rate of decomposition is constant.

(2) Double compartment models: These models assume that incoming plant material

is divided into two pools, a fast and a slow pool, which decompose according to first

order kinetics. Mathematically this can be written as : C = Cs (1-e-st) + Cf (1-e-ft), where

C represents total C mineralized over a given time period t, Cs and Cf are the stable and

labile fractions of organic C, respectively, and s and f are the proportional rate

constants for C mineralization for the stable and labile fractions, respectively.

(3) Non-compartmental decay models: The assumption behind these models is that

during decomposition the organic matter moves down a quality scale. The quality of

organic matter varies between zero for resistant plant material and one for fresh

organic material.

(4) Multi-compartmental models: These models assume that SOC is decomposed into

various pools which in turn, decay by first order kinetics. Examples are CENTURY

(Parton et al., 1988), RothC (Jenkinson, 1990), DAISY (Jensen et al., 1997) and DNDC (Li

et al., 1997).

19

Models in categories 1 and 2 are static (environmental variables remain constant),

whereas the models in categories 3 and 4 are dynamic (environmental variables vary

with time) (Falloon and Smith, 2000). The dynamic models can be further split into

following four broad classes (Paustian, 1994):

(1) Process-orientated models: simulate the processes controlling the flow of energy

and matter transformations. Hence, SOC pools are linked to each other by C flow from

one pool to another over time and these pools have different turnover time. The

models in this category include CENTURY (Parton et al., 1988), RothC (Jenkinson,

1990), DAISY (Jensen et al., 1997), CANDY (Franko et al., 1997) and DNDC (Li et al.,

1997) etc.

(2) Organism-orientated models: quantify the flow of nutrients and energy in soil

biota, through various functional or taxonomic groups. For example: Darrah (a model

for Rhizosphere C flow, Darrah, 1991).

(3) Integrated models: link process models to organism-orientated models. For

example: SOMM (Soil organic matter dynamics in relation to decomposer organisms,

Chertov and Komarov, 1996).

(4) Cohort models: convert SOM into different cohorts which are further divided into

pools. Unlike process-orientated models where the SOM decomposition is regulated

by physical and biochemical processes, these models consider microbial physiology as

the main driving factor for SOM decomposition (Batlle-Aguilar et al., 2010). For

example: SOMCO (soil organic matter cohort, Gignoux et al., 2001).

To date, there are 250 models for predicting SOC dynamics and nutrient

turnover (Manzoni and Porporato, 2009). These models differ in underlying

assumptions and the processes responsible for SOM decomposition. Nevertheless,

most of these SOM models divide SOC into different pools with distinct properties and

rate constants. Another class of models determine SOC decomposition by an attribute

called “quality”, the so-called Q model (Bosatta and Agren 1985) where the rate of

SOM decomposition is associated with a continuous change in “quality”.

20

1.10 Evaluation, comparison and application of soil organic carbon models

Once a model has been developed, its performance needs to be evaluated to:

1. assess how well the model performs in the situation for which it was

developed.

2. understand the reasons if the model fails to perform well in a given situation.

3. predict future changes in SOC with confidence and a maximum degree of

accuracy.

Models can be evaluated at point, regional or global scale. The performance of

models depends upon required inputs, availability of input data at appropriate scale

and hypotheses made for developing the model. Smith et al. (1997) evaluated nine

SOC models against data from seven long term experiments under arable, grassland

and woodland. Their results showed that six models (RothC, DNDC, CANDY, CENTURY,

DAISY and NCSOIL) performed well with no significant difference in accuracy; the

residual mean square error of these models was lower than the other group of models

(ITE, Verberne and SOMM). Similarly, five SOC models (RothC, DNDC, CENTURY, EPIC

and SOCRATES) were evaluated by Izaurralde et al. (1996) at a site specific and a

regional level in Canada. Among these models, SOCRATES was best for simulating long

term SOC trends at site specific levels, but CENTURY was superior in predicting SOC

trends at regional level because it, unlike SOCRATES, allows for different management

strategies. Among all models, CENTURY (Parton et al., 1988) and RothC (Jenkinson et

al., 1987) are most widely used for studying the impact of climate and land-use change

on SOC.

CENTURY was originally developed for grasslands (Parton et al., 1988), but it

has also been used to simulate SOC in forest (Kelley et al., 1997; Kirschbaum and Paul,

2002) and cropping (Carter et al., 1993) systems. Parton et al. (1993) evaluated

CENTURY using data from 11 temperate and tropical grasslands around the world. The

results showed that SOC and N levels could be simulated to within ±25% of the

observed values for a diverse set of soils. Kelley et al. (1997) used CENTURY to

accurately predict SOC in arable land (Germany, Czech Republic and Australia),

grassland (UK), forest (USA) and wood land (UK). It has also been successfully used at

field scale in Kenya (Cerri et al. 2007) and Denmark (Foereid and Hogh-Jensen 2004),

and for grassland converted to continuous fallow in Russia (Mikhailova et al. 2000).

21

Using RothC, simulated SOC contents were an acceptable approximation of

measured SOC in long-term experiments in Germany, Australia, Czech Republic,

England and the USA (Coleman et al., 1997). RothC has also been successfully applied

in China (Guo et al., 2007), Kenya (Kamoni et al., 2007), Zambia (Kaonga and Coleman,

2008) and West Africa (Nakamura et al., 2010). Gottschalk et al. (2010) found that the

ability of RothC to simulate SOC after deforestation can be improved after

incorporating a new SOC pool (labile but protected carbon fraction). Cerri et al. (2003,

2007) used a chronosequence approach for studying changes in SOC after conversion

of forest to pasture in the Brazilian Amazon and showed that RothC accurately

simulated a decline in SOC following clearing and conversion to pasture.

1.11 Rothamsted carbon model

Among the SOC models, RothC (Jenkinson and Rayner 1977, Jenkinson et al.

1987, Coleman and Jenkinson 1996), is being widely used worldwide and is also used

for carbon accounting in Australia (Skjemstad et al., 2004). RothC includes five pools of

SOM: decomposable plant material (DPM, labile – cellulose and hemicellulose),

resistant plant material (RPM, recalcitrant – lignin), microbial biomass (BIO, soil

microbes), humidified SOM (HUM, humus) and inert organic matter (IOM) that is

resistant to biological transformations (Coleman and Jenkinson, 1996). The incoming

plant material is divided into DPM and RPM but the DPM/RPM ratio depends upon the

vegetation type (1.44 for crops and grasses, 0.25 for deciduous or tropical woodlands

and 0.67 for unimproved grassland and scrubland). Both DPM and RPM decompose to

form CO2, BIO and HUM (Figure 4). Biomass and humus further decompose into CO2,

biomass and humus pools but at a slower rate than DPM and RPM.

The amount of C (y) lost from each pool, except IOM, is described by first order

exponential decay:

y = y0 (1-e-abckt)

where y0 = initial amount of C in a particular pool; a, b and c are rate modifying factors

for temperature, moisture and plant retainment, respectively; k is the rate constant

for the given pool expressed per year and t is 1/12 to convert k to a monthly time-

step.

22

Figure 4: Structure of the RothC model

The rate modifying factor ‘b’ accounts for the soil moisture deficit which is a

function of soil clay content, monthly average rainfall and pan evaporation.

Decomposition in the RothC model is also sensitive to whether the soil is bare or

covered by vegetation. The modifier is 0.6 for actively growing vegetation (reduces

decomposition rates by 40%) and 1 for bare soil. The pools are defined by the rate

constant, proportion of BIO to HUM (fixed in the model) and the ratio of CO2 to BIO +

HUM which varies according to the clay content with lower clay content leading to

higher relative CO2 release. The rate constant ‘k’ is 10 yr-1 for DPM, 0.30 yr-1 for RPM,

0.66 yr-1 for BIO and 0.02 yr-1 for HUM.

The inputs required to run the model are divided into three types:

1. Soil data : Initial SOC ( t C ha-1), clay (%), depth of soil layer sampled (cm)

2. Land use and management data: soil cover, monthly input of plant residue (t

ha-1), and farm yard manure ( t ha-1).

3. Climate data: Monthly rainfall (mm), monthly open pan evaporation (mm) and

average monthly mean air temperature (°C)

These input parameters must be known to run the model but plant input is rarely

known and can be generated by the model by running it in inverse mode.

IOM

CO2

Organic inputs

DPM

RPM

HUM

BIO

HUM

BIO

CO2

23

1.12 Modifications of RothC for subsoils, waterlogged soils and Andosols

According to Coleman and Jenkinson (2005), RothC should be used cautiously

on subsoils, soil from tundra and taiga regions, soils on recent volcanic ash and is not

suitable for waterlogged soils. However, Jenkinson and Coleman (2008) also

parameterized RothC for subsoils after incorporating two parameters, p (for downward

movement of organic C) and s (for slow decomposition of SOM at depth). Shirato et al.

(2004) modified RothC for Andosols of Japan by changing the rate constant of humic

pool and setting inert organic matter to zero. They found that with the modified

RothC, there was a good agreement between measured and modelled SOC of long-

term experiments on these soils. Further, Shirato et al. (2005) modified the

decomposition rate constants of RothC by a factor of 0.2 in summer and 0.6 in winter

for better performance of this model for paddy soils of Japan. These studies suggest

that RothC can be parameterized for different conditions by adjusting rate constants or

adding additional parameters.

1.13 Measurable and modelled pools in RothC

The input pools in the model can be approximated to measurable pools either

by chemical analyses or decomposition studies of different plant residues. Smith et al.

(2002) showed that a measured pool is equivalent to a modelled pool only when it is

unique and non-composite. Most studies initialize RothC using only measured total soil

carbon and other input parameters, however there are few studies where the

conceptual pool of model were replaced with measured SOC pools. Skjemstad et al.

(2004) showed that RPM, HUM and IOM pools of RothC could be replaced by POC (C

associated with particles > 53 µm), humus (C associated with particles < 53 µm

excluding char-C) and char-C (condensed aromatic C), respectively. They found a good

agreement between measured and modelled pools after changing the decomposition

rate constant (k) of RPM from 0.30 to 0.15 per year and retaining the original rate

constant values for the other pools as proposed by Jenkinson et al. (1987). Similarly,

Zimmermann et al. (2006) used a fractionation scheme to divide SOC pools into five

fractions namely POC, DOC, C associated with silt plus clay (SSOC), C associated with

sand and stable aggregates (ASOC), and resistant C (RSOC). They found that POC+DOC

could be replaced with DPM + RPM, SSOC (minus RSOC) and ASOC with HUM+BIO and RSOC

24

with IOM. Shirato and Yokozawa (2006) identified only DPM and RPM fractions in

RothC by acid hydrolysis of plant materials. They divided C in plant material into three

pools: labile pool I (obtained by hydrolysis with 5 N H2SO4), labile pool II (obtained by

hydrolysis with 26 N H2SO4 and then with 2N H2SO4) and recalcitrant pool

(unhydrolyzed residue). They found that the labile pool approximated the DPM pool

and the labile pool II plus recalcitrant pool approximated the RPM pool in RothC.

Ludwig et al. (2003) compared measured and modelled pools of RothC using the 13C

technique. Total C and maize derived C in the <63 µm fraction were correlated with the

sum of modelled total and maize derived C in the humic pool, inert organic matter and

microbial biomass.

1.14 Integration of RothC with spatial data in the geographical information system

For identification of potential areas for C sequestration at regional scale,

information about spatial variation in climate, soil properties, vegetation, crops and

land use is required (Falloon et al., 1998; Smith et al., 2006). Many SOC models are

point-based and perform simulations of SOC and predictions for one site at a time. The

integration of the RothC with soil, land use and climate data in a GIS environment was

successfully illustrated by Fallon et al. (1998) for studying the changes in SOC after 50

and 100 years following afforestation of arable land in Hungary. Fallon et al. (2006)

studied SOC fluxes in mineral soils as a function of changes in climate, land use and

land management at a 1-km resolution in the UK. Smith et al. (2005, 2006) used a

similar approach for studying projected changes in SOC of European croplands,

grassland and forests in 18 x 18 km cells. Further, Smith et al. (2007) used the spatial

version of RothC to estimate changes in SOC stocks in soils of European Russia and the

Ukraine from 1990-2070 with four intergovernmental panel on climate change (IPCC)

scenarios. Similarly, Kamoni et al. (2007), Al-Admat et al. (2007), Bhattacharya et al.

(2007) and Cerri et al. (2007) predicted SOC stocks between 2000 and 2030 in Kenya,

Jordan, India and Brazilian Amazon, respectively. Jones et al. (2005) used RothC to

model the effect of future climate on global SOC stocks and predicted that SOC stocks

would remain unchanged until about 2050 and but decline sharply thereafter, resulting

in a decrease of SOC by 54 Pg C by 2100. In Australia, FullCAM (Richards, 2001) is used

for C accounting and can also be run in a spatial mode for studying the effect of land

25

use changes after integrating remote sensing data, climatic parameters and soil

information. FullCAM is an integration of several sub-models: RothC and CENTURY, the

empirical C tracking model CAMFor (Richards and Evans, 2000), the tree growth model

3PG (Landsberg and Waring, 1997) and the litter decomposition model GENDEC

(Moorhead et al., 1999).

1.15 Conclusions and knowledge gaps

In South Australia, dryland salinity and dry saline land salinity are forms of land

degradation of major significance, but information about the extent of the area

affected is scant. By providing fast, cost effective and time series data, remote sensing

can play an important role in detecting, mapping, and monitoring salt-affected surface

features. In this thesis, the ability of higher resolution multispectral satellite imagery as

a means of mapping salinity at a large scale will be assessed in the agricultural area

around Kadina, South Australia where dryland and dry land salinity are wide-spread.

Increasing soil salinity and sodicity affect SOC dynamics and can alter C stocks and

fluxes in the landscape, but the influence of salt on SOC stocks may vary according to

different types of salinity and climatic conditions. Therefore, an agricultural area in

Punjab (Muktsar district), India where salinity was induced by irrigation, will also

mapped and sampled on a 1:50,000 scale.

Although there are many studies on the effect of EC and SAR on SOM

decomposition, they mainly used soils to which salt was added. This may not reflect

the effect of EC and SAR on decomposition in soils that have been exposed to salt for

longer periods of time. To close this knowledge gap, SOM decomposition in field

collected salt-affected and salt amended soils (salinity is altered by adding soluble

salts) will be determined.

Compared to other models, RothC is quite simple and transparent model and

has been used successfully to predict changes in SOC in different parts of the world.

However, RothC has not been calibrated for salt-affected soils that cover large areas in

Australia and India. In this thesis, an approach will be described for development of a

decomposition rate modifier for salinity and its integration into the spatial version of

RothC. Further, the impact of reduced plant inputs in salt-affected soils on modelled

26

SOC stocks will also be evaluated. Lastly, the RothC, modified for salt-affected soils, will

be used to model regional SOC stocks.

Thus, the present study has the following aims:

1. to map salt-affected soils of selected regions of South Australia and India

2. to investigate the effect of various levels of EC and SAR on SOC turnover

3. to develop decomposition rate modifier for salt that can be included into

RothC for simulating SOC dynamics in salt-affected soils

4. to estimate regional SOC turnover in salt-affected soils by linking the

modified Roth C model to a GIS data base for India and Australia

1.16 Structure of thesis

A schematic diagram of outline of approaches used for studying SOC dynamics

in salt-affected soils by integrating remote sensing, incubation experiments, SOC

modelling and GIS is shown in Figure 5. For this purpose, two regions, one in India

(Muktsar, Punjab) and one in Australia (Kadina, South Australia) were selected. The

salt-affected soils were mapped on finer scale (~ 1:10, 000) for Kadina, South Australia

(Chapter 2) and on coarse scale (1:50,000 scale) for Muktsar, Punjab, India (Appendix

1) due to best available data.

The salt-affected soils contain calcium carbonate and estimation of CO2 release

in calcareous soils is problematic. Chapter 3 covers the influence of calcium carbonate

on CO2 release in saline soils in the presence of residues. Previous studies (Ben-Gal et

al., 2009; Rengasamy, 2010) suggested that the osmotic potential of soil solution may

be an appropriate parameter for assessing the effect of salinity on plant growth. The

question arose whether the effect of salinity on soil respiration may be different when

calculated with osmotic potential. Therefore, salt-amended soils of varying texture

were used in an incubation study to assess the effect of salinity on CO2 release as a

function of osmotic potential (Chapter 4). The results of this study showed that salt-

amended soils overestimate the salinity effect compared to saline soils from the field.

Chapter 5 covers the relationship between CO2 release and soil properties in salt-

affected landscapes of India and Australia. The results of this study suggest that among

27

all soil properties, EC was the main factor influencing for soil respiration. Therefore, EC

should be taken account into simulating SOC dynamics in these soils.

Currently SOC models do not take account into salinity and the plant growth is

lower in salt-affected soils. Chapter 6 covers the development of decomposition rate

modifier for salinity using CO2 release from soils of India and Australia. RothC, modified

after including a decomposition rate modifier and a plant input modifier, was used to

simulate SOC dynamics in soils collected from a field in South Australia. In order to

estimate the past SOC content from saline soils of India and Australia when these were

non-saline, a new method to model the change in SOC stocks that can be attributed to

the salinization of the soils was developed (Chapter 7) using RothC modified for saline

soils.

28

Figure 5: Schematic diagram of the structure of thesis

Australia

Calcium carbonate

Salt amended soils

Soil properties

(Chapter 3)

(Chapter 4)

(Chapter 5)

India

Past, present and future SOC stocks (India and Australia)

Remote Sensing (Satellite imagery)

Classification

Selection of soil sampling sites

Measurement of total soil organic carbon, EC and SAR

Incubation experiments for measuring CO2 emission

Effect of salt on CO2 emission

Compared with modelled CO2 (RothC) emission

Salt rate modifier for decomposition

RothC modified for saline soils Plant input modifier for salinity

(Chapter 2)

(Chapter 6)

(Chapter 7)

29

References

Abdou, F., El-Kobbia, T., Sorensen, L., 1975. Decomposition of native organic matter and 14C-labelled barley straw in different Egyptian soils. Beiträge zur Tropischen Landwirtschaft und Veterinärmedizin 13 203-209. Agarwal, A.S., Singh, B.R., Kanehiro, Y., 1971. Ionic effect of salts on mineral nitrogen release in an allophanic soil. Soil Science Society of America Journal 35, 454. Al-Adamat, R., Rawajfih, Z., Easter, M., Paustian, K., Coleman, K., Milne, E., Falloon, P., Powlson, D., Batjes, N., 2007. Predicted soil organic carbon stocks and changes in Jordan between 2000 and 2030 made using the GEFSOC Modelling System. Agriculture, Ecosystems and Environment 122, 35-45. Amundson, R., 2001. The carbon budget in soils. Annual Review of Earth and Planetary Sciences 29, 535-562. Ardö, J., Olsson, L., 2003. Assessment of soil organic carbon in semi-arid Sudan using GIS and the CENTURY model. Journal of Arid Environments 54, 633-651. Bajwa, M., Josan, A., Hira, G., Singh, N., 1986. Effect of sustained saline irrigation on soil salinity and crop yields. Irrigation Science 7, 27-35. Baldock, J., Nelson, P., 1999. Soil organic matter. Handbook of Soil Science CRC Press, Boca Raton, FL, pp. D3-D18. Baldock, J., Skjemstad, J., 1999. Soil organic carbon/soil organic matter. Soil analysis: An interpretation manual. CSIRO Publishing, Melbourne, Australia, 159–170. Bannari, A., Guedon, A., El-Harti, A., Cherkaoui, F., El-Ghmari, A., 2008. Characterization of slightly and moderately saline and sodic soils in irrigated agricultural land using simulated data of Advanced Land Imaging (EO-1) Sensor. Communications in Soil Science and Plant Analysis 39, 2795-2811. Batlle-Aguilar, J., Brovelli, A., Porporato, A., Barry, D., 2010. Modelling soil carbon and nitrogen cycles during land use change. A review. Agronomy for Sustainable Development 10.1051/agro/2010007. Ben-Gal, A., Borochov-Neori, H., Yermiyahu, U., Shani, U., 2009. Is osmotic potential a more appropriate property than electrical conductivity for evaluating whole-plant response to salinity? Environmental and Experimental Botany 65, 232-237. Bhattacharyya, T., Pal, D., Easter, M., Batjes, N., Milne, E., Gajbhiye, K., Chandran, P., Ray, S., Mandal, C., Paustian, K., 2007. Modelled soil organic carbon stocks and changes in the Indo-Gangetic Plains, India from 1980 to 2030. Agriculture, Ecosystems and Environment 122, 84-94.

30

Bosatta, E., Agren, G.I., 1985. Theoretical analysis of decomposition of heterogeneous substrates. Soil Biology and Biochemistry 17, 601-610 Brady, N.C., Weil, R.R., 1999. The nature and properties of soils. Prentice Hall Upper Saddle River, NJ. Broadbent, F., Nakashima, T., 1971. Effect of added salts on nitrogen mineralization in three California soils. Soil Science Society of America Journal 35, 457-460. Cahyani, V., Watanabe, A., Matsuya, K., Asakawa, S., Kimura, M., 2002. Succession of microbiota estimated by phospholipid fatty acid analysis and changes in organic constituents during the composting process of rice straw. Soil Science and Plant Nutrition 48, 735-743. Carter, M., Parton, W., Rowland, I., Schultz, J., Steed, G., 1993. Simulation of soil organic carbon and nitrogen changes in cereal and pasture systems of Southern Australia. Australian Journal of Soil Research 31, 481-491. Cerri, C., Coleman, K., Jenkinson, D., Bernoux, M., Victoria, R., Cerri, C., 2003. Modeling soil carbon from forest and pasture ecosystems of Amazon, Brazil. Soil Science Society of America Journal 67, 1879-1887. Cerri, C, Easter, M., Paustian, K., Killian, K., Coleman, K., Bernoux, M., Falloon, P., Powlson, D.S., Batjes, N., Milne, E., 2007. Simulating SOC changes in 11 land use change chronosequences from the Brazilian Amazon with RothC and Century models. Agriculture, Ecosystems and Environment 122, 46-57. Chandra, S., Joshi, H., Pathak, H., Jain, M., Kalra, N., 2002. Effect of potassium salts and distillery effluent on carbon mineralization in soil. Bioresource Technology 83, 255-257. Chertov, O., Komarov, A., 1996. SOMM-a model of soil organic matter and nitrogen dynamics in terrestrial ecosystems. NATO ASI SERIES 38, 231-236. Coleman, K., Jenkinson, D., 1996. RothC-26.3-A Model for the turnover of carbon in soil. In: Powlson, D., Smith, P., Smith, J. (Eds.), Evaluation of soil organic matter models using existing long-term datasets. NATO ASI Series 38, 237-246. Coleman, K., Jenkinson, D., 2005. RothC 26.3 – A model for the turnover of carbon in soil: Model description and users guide. Lawes Agricultural Trust, Harpenden. Coleman, K., Jenkinson, D., Crocker, G., Grace, P., Klir, J., Körschens, M., Poulton, P., Richter, D., 1997. Simulating trends in soil organic carbon in long-term experiments using RothC-26.3. Geoderma 81, 29-44. Darrah, P., 1991. Models of the rhizosphere. I. Microbial population dynamics around a root releasing soluble and insoluble carbon. Plant and Soil 133, 187-199.

31

Dendooven, L., Alcántara-Hernández, R.J., Valenzuela-Encinas, C., Luna-Guido, M., Perez-Guevara, F., Marsch, R., 2010. Dynamics of carbon and nitrogen in an extreme alkaline saline soil: A review. Soil Biology and Biochemistry 42, 865-877. Dutkiewicz, A., Lewis, M., Ostendorf, B., 2009. Evaluation and comparison of hyperspectral imagery for mapping surface symptoms of dryland salinity. International Journal of Remote Sensing 30, 693-719. Dutkiewicz, A., Lewisa, M., Ostendorf, B., 2006. Mapping surface symptoms of dryland salinity with hyperspectral imagery. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences Vol. 34, Part XXX [CD-ROM] Dwivedi, R., 1996. Monitoring of salt-affected soils of the Indo-Gangetic alluvial plains using principal component analysis. International Journal of Remote Sensing 17, 1907-1914. Dwivedi, R., Kothapalli, R., Singh, A., 2009. Generation of farm-level information on salt affected soils using IKONOS-II multispectral data. In: Metternicht, G., Zinck, J. (Eds.), Remote Sensing of soil salinization : Impact on land management. CRC Press, New York, pp. 73-90. Evans, F., Caccetta, P., 2000. Broad-scale spatial prediction of areas at risk from dryland salinity. Cartography 29, 33-40. Falloon, P., Smith, P., 2000. Modelling refractory soil organic matter. Biology and Fertility of Soils 30, 388-398. Falloon, P., Smith, P., Bradley, R., Milne, R., Tomlinson, R., Viner, D., Livermore, M., Brown, T., 2006. RothCUK–a dynamic modelling system for estimating changes in soil C from mineral soils at 1 km resolution in the UK. Soil use and management 22, 274-288. Falloon, P., Smith, P., Smith, J., Szabo, J., Coleman, K., Marshall, S., 1998. Regional estimates of carbon sequestration potential: linking the Rothamsted Carbon Model to GIS databases. Biology and Fertility of Soils 27, 236-241. Foereid, B., Hogh-Jensen, H., 2004. Carbon sequestration potential of organic agriculture in northern Europe–a modelling approach. Nutrient Cycling in Agroecosystems 68, 13-24. Franko, U., Crocker, G., Grace, P., Klir, J., Körschens, M., Poulton, P., Richter, D., 1997. Simulating trends in soil organic carbon in long-term experiments using the CANDY model. Geoderma 81, 109-120. Garcia, C., Hernandez, T., 1996. Influence of salinity on the biological and biochemical activity of a calciorthird soil. Plant and Soil 178, 255-263.

32

Ghassemi, F., Jakeman, A.J., Nix, H.A., 1995. Salinisation of Land and Water Resources: human causes, extent, management and case studies. Cab International, Wallingford, UK. Ghollarata, M., Raiesi, F., 2007. The adverse effects of soil salinization on the growth of Trifolium alexandrinum L. and associated microbial and biochemical properties in a soil from Iran. Soil Biology and Biochemistry 39, 1699-1702. Ghoulam, C., Foursy, A., Fares, K., 2002. Effects of salt stress on growth, inorganic ions and proline accumulation in relation to osmotic adjustment in five sugar beet cultivars. Environmental and Experimental Botany 47, 39-50. Gignoux, J., House, J., Hall, D., Masse, D., Nacro, H.B., Abbadie, L., 2001. Design and test of a generic cohort model of soil organic matter decomposition: the SOMKO model. Global Ecology and Biogeography 10, 639-660. Goldshleger, N., Ben Dor, E., Benyamini, Y., Agassi, M., Blumberg, D., 2001. Characterization of soil’s structural crust by spectral reflectance in the SWIR region (1.2–2.5 m). Terra Nova 13, 12-17. Gottschalk, P., Bellarby, J., Chenu, C., Foereid, B., Smith, P., Wattenbach, M., Zingore, S., Smith, J., 2010. Simulation of soil organic carbon response at forest cultivation sequences using 13C measurements. Organic Geochemistry 41, 41-54. Gros, R., Poly, F., Jocteur Monrozier, L., Faivre, P., 2003. Plant and soil microbial community responses to solid waste leachates diffusion on grassland. Plant and Soil 255, 445-455. Guo, L., Falloon, P., Coleman, K., Zhou, B., Li, Y., Lin, E., Zhang, F., 2007. Application of the RothC model to the results of long term experiments on typical upland soils in northern China. Soil Use and Management 23, 63-70. Howari, F., 2003. The use of remote sensing data to extract information from agricultural land with emphasis on soil salinity. Australian Journal of Soil Research 41, 1243-1253. Isbell, R.F., 2002. The Australian soil classification. CSIRO Publishing, Collingwood, Australia. Izaurralde, R., McGill, W., Jans-Hammermeister, D., Haugen-Korzyra, K., Hiley, J., 1996. Development of a technique to calculate carbon fluxes in agricultural soils at the ecodistrict level using simulation models and various aggregation methods. Interim Report, Submitted to the Scientific Authority of the Agriculture and Agri-Food Canada Greenhouse Gas Initiative, University of Alberta, Edmonton, AB.

33

Jastrow, J.D., Amonette, J.E., Bailey, V.L., 2007. Mechanisms controlling soil carbon turnover and their potential application for enhancing carbon sequestration. Climatic Change 80, 5-23. Jenkinson, D., 1990. The turnover of organic carbon and nitrogen in soil. Philosophical Transactions: Biological Sciences 329, 361-368. Jenkinson, D., Coleman, K., 2008. The turnover of organic carbon in subsoils. Part 2. Modelling carbon turnover. European Journal of Soil Science 59, 400-413. Jenkinson, D., Hart, P., Rayner, J., Parry, L., 1987. Modelling the turnover of organic matter in long-term experiments at Rothamsted. Intecol. Bull 15, 1-8. Jenkinson, D., Rayner, J., 1977. The turnover of soil organic matter in some of the Rothamsted classical experiments. Soil Science 123, 298. Jensen, L., Mueller, T., Nielsen, N., Hansen, S., Crocker, G., Grace, P., Klir, J., Körschens, M., Poulton, P., 1997. Simulating trends in soil organic carbon in long-term experiments using the soil-plant-atmosphere model DAISY1. Geoderma 81, 5-28. Johnston, R., Barson, M., 1990. An assessment of the use of remote sensing techniques in land degradation studies, Bulletin 5. Australian Department of Primary Industries and Energy, Canberra. Jones, C., McConnell, C., Coleman, K., Cox, P., Falloon, P., Jenkinson, D., Powlson, D., 2005. Global climate change and soil carbon stocks; predictions from two contrasting models for the turnover of organic carbon in soil. Global Change Biology 11, 154-166. Joshi, M., Sahai, B., 1993. Mapping of salt-affected land in Saurashtra coast using Landsat satellite data. International Journal of Remote Sensing 14, 1919-1929. Kalra, N., Joshi, D., 1996. Potentiality of Landsat, SPOT and IRS satellite imagery, for recognition of salt-affected soils in Indian arid zone. International Journal of Remote Sensing 17, 3001-3014. Kamoni, P., Gicheru, P., Wokabi, S., Easter, M., Milne, E., Coleman, K., Falloon, P., Paustian, K., Killian, K., Kihanda, F., 2007. Evaluation of two soil carbon models using two Kenyan long term experimental datasets. Agriculture, Ecosystems and Environment 122, 95-104. Kaonga, M., Coleman, K., 2008. Modelling soil organic carbon turnover in improved fallows in eastern Zambia using the RothC-26.3 model. Forest Ecology and Management 256, 1160-1166. Kaur, B., Gupta, S., Singh, G., 2002. Bioamelioration of a sodic soil by silvopastoral systems in northwestern India. Agroforestry Systems 54, 13-20.

34

Kelly, R., Parton, W., Crocker, G., Graced, P., Klir, J., Körschens, M., Poulton, P., Richter, D., 1997. Simulating trends in soil organic carbon in long-term experiments using the CENTURY model. Geoderma 81, 75-90. Killham, K., 1994. Soil ecology. Cambridge University Press. Kirschbaum, M.U.F., Paul, K.I., 2002. Modelling C and N dynamics in forest soils with a modified version of the CENTURY model. Soil Biology and Biochemistry 34, 341-354. Lal, R., 2001. Potential of desertification control to sequester carbon and mitigate the greenhouse effect. Climatic Change 51, 35-72. Lal, R., 2004. Soil carbon sequestration to mitigate climate change. Geoderma 123, 1-22. Lal, R., 2008. Carbon sequestration. Philosophical Transactions of the Royal Society B: Biological Sciences 363, 815-830. Landsberg, J., Waring, R., 1997. A generalised model of forest productivity using simplified concepts of radiation-use efficiency, carbon balance and partitioning. Forest Ecology and Management 95, 209-228. Laura, R.D., 1974. Effects of neutral salts on carbon and nitrogen mineralization of organic matter in soil. Plant and Soil 41, 113-127. Laura, R.D., 1977. Salinity and nitrogen mineralization in soil. Soil Biology and Biochemistry 9, 333-336. Lewis, M., 2002. Spectral characterization of Australian arid zone plants. Canadian journal of remote sensing 28, 219-230. Li, C., Frolking, S., Crocker, G., Grace, P., Klír, J., Körchens, M., Poulton, P., 1997. Simulating trends in soil organic carbon in long-term experiments using the DNDC model. Geoderma 81, 45-60. Li, X., Li, F., Singh, B., Cui, Z., Rengel, Z., 2006a. Decomposition of maize straw in saline soil. Biology and Fertility of Soils 42, 366-370. Li, X., Li, F., Ma, Q., Cui, Z., 2006b. Interactions of NaCl and Na2SO4 on soil organic C mineralization after addition of maize straws. Soil Biology and Biochemistry 38, 2328-2335. Liu, W., Kogan, F., 2002. Monitoring Brazilian soybean production using NOAA/AVHRR based vegetation condition indices. International Journal of Remote Sensing 23, 1161-1179.

35

Ludwig, B., John, B., Ellerbrock, R., Kaiser, M., Flessa, H., 2003. Stabilization of carbon from maize in a sandy soil in a long term experiment. European Journal of Soil Science 54, 117-126. Manchanda, M., 1984. Use of remote sensing techniques in the study of distribution of salt-affected soils in North-West India. Journal of the Indian Society of Soil Science 32, 701-706. Manzoni, S., Porporato, A., 2009. Soil carbon and nitrogen mineralization: theory and models across scales. Soil Biology and Biochemistry 41, 1355-1379. Marschner, H., 1995. Mineral nutrition of higher plants. Academic press London. Mass, E., Hoffman, G., 1977. Crop salt tolerance current assessment. Journal of the Irrigation and Drainage Division ASCE 504, 115-135. McClung, G., Frankenberger, W., 1987. Nitrogen mineralization rates in saline vs. salt-amended soils. Plant and Soil 104, 13-21. McCormick, R.W., Wolf, D.C., 1980. Effect of sodium chloride on CO2 evolution, ammonification, and nitrification in a Sassafras sandy loam. Soil Biology and Biochemistry 12, 153-157. Meloni, D., Oliva, M., Ruiz, H., Martinez, C., 2001. Contribution of proline and inorganic solutes to osmotic adjustment in cotton under salt stress. Journal of Plant Nutrition 24, 599-612. Metternicht, G., Zinck, J., 1997. Spatial discrimination of salt-and sodium-affected soil surfaces. International Journal of Remote Sensing 18, 2571-2586. Metternicht, G., Zinck, J., 1998. Evaluating the information content of JERS-1 SAR and Landsat TM data for discrimination of soil erosion features. ISPRS Journal of Photogrammetry and Remote Sensing 53, 143-153. Metternicht, G., Zinck, J., 2003. Remote sensing of soil salinity: potentials and constraints. Remote Sensing of Environment 85, 1-20. Metternicht, G., Zinck, J.A., 2009. Remote sensing of soil salinization: Impact on land management. CRC, Boca Raton, FL. Mikhailova, E., Bryant, R., DeGloria, S., Post, C., Vassenev, I., 2000. Modeling soil organic matter dynamics after conversion of native grassland to long-term continuous fallow using the CENTURY model. Ecological Modelling 132, 247-257. Moorhead, D., Currie, W., Rastetter, E., Parton, W., Harmon, M., 1999. Climate and litter quality controls on decomposition: an analysis of modeling approaches. Global Biogeochemical Cycles 13, 575-589.

36

Nakamura, S., Hayashi, K., Omae, H., Ramadjita, T., Dougbedji, F., Shinjo, H., Saidou, A.K., Tobita, S., 2010. Validation of soil organic carbon dynamics model in the semi-arid tropics in Niger, West Africa. Nutrient Cycling in Agroecosystems, doi 10.1007/s10705-010-9402-4. Nelson, P., Baldock, J., Clarke, P., Oades, J., Churchman, G., 1999. Dispersed clay and organic matter in soil: their nature and associations. Australian Journal of Soil Research 37, 289-315. Nelson, P.N., Ladd, J.N., Oades, J.M., 1996. Decomposition of 14C-labelled plant material in a salt-affected soil. Soil Biology and Biochemistry 28, 433-441. Oren, A., 1999. Bioenergetic aspects of halophilism. Microbiology and molecular biology reviews 63, 334-348. Pankhurst, C., Yu, S., Hawke, B., Harch, B., 2001. Capacity of fatty acid profiles and substrate utilization patterns to describe differences in soil microbial communities associated with increased salinity or alkalinity at three locations in South Australia. Biology and Fertility of Soils 33, 204-217. Parihar, J., Panigrahy, S., Sharma, S., Patnaik, C., Dutta, S., 2009. GEOSS Ag 07 03 discussion paper. Conference on "Impact of climate change on agriculture", Ahmedabad, India. Parton, W., 1996. The CENTURY model. In: Powlson, D., Smith, P., Smith, J. (Eds.), Evaluation of soil organic matter models using existing long-term datasets. NATO ASI Series 38, 283-294. Parton, W., Scurlock, J., Ojima, D., Gilmanov, T., Scholes, R., Schimel, D., Kirchner, T., Menaut, J., Seastedt, T., Moya, E.G., 1993. Observations and modeling of biomass and soil organic matter dynamics for the grassland biome worldwide. Global Biogeochemical Cycles 7, 785-809. Parton, W.J., Stewart, J.W.B., Cole, C.V., 1988. Dynamics of C, N, P and S in grassland soils: a model. Biogeochemistry 5, 109-131. Pathak, H., Rao, D.L.N., 1998. Carbon and nitrogen mineralization from added organic matter in saline and alkali soils. Soil Biology and Biochemistry 30, 695-702. Paustian, K., 1994. Modelling soil biology and biochemical processes for sustainable agriculture research. In: Paustian, K., Pankhurst, C., Doube, B., Gupta, V., Grace, P. (Eds.), Soil biota: management in sustainable farming systems. CSIRO Publications, pp. 182-193. Pessarakli, M., Szabolcs, I., 1999. Soil salinity and sodicity as particular plant/crop stress factors. In: Pessarakli, M. (Ed.), Handbook of Plant and Crop Stress. Marcel Dekker, New York, pp 1–16.

37

Quirk, J., 2001. The significance of the threshold and turbidity concentrations in relation to sodicity and microstructure. Australian Journal of Soil Research 39, 1185-1217. Rantalainen, M.L., Kontiola, L., Haimi, J., Fritze, H., Setälä, H., 2004. Influence of resource quality on the composition of soil decomposer community in fragmented and continuous habitat. Soil Biology and Biochemistry 36, 1983-1996. Rao, B., Dwivedi, R., Venkataratnam, L., Ravishankar, T., Thammappa, S., Bhargawa, G., Singh, A., 1991. Mapping the magnitude of sodicity in part of the Indo-Gangetic plains of Uttar Pradesh, Northern India using Landsat-TM data. International Journal of Remote Sensing 12, 419-425. Rasul, G., Appuhn, A., Müller, T., Joergensen, R.G., 2006. Salinity-induced changes in the microbial use of sugarcane filter cake added to soil. Applied Soil Ecology 31, 1-10. Rengasamy, P., 2002. Transient salinity and subsoil constraints to dryland farming in Australian sodic soils: an overview. Australian Journal of Experimental Agriculture 42, 351-361. Rengasamy, P., 2006. World salinization with emphasis on Australia. Journal of Experimental Botany 57, 1017. Rengasamy, P., 2010. Soil processes affecting crop production in salt-affected soils. Functional Plant Biology 37, 613-620. Rengasamy, P., Olsson, K., 1991. Sodicity and soil structure. Australian Journal of Soil Research 29, 935-952. Richards, G., Evans, D., 2000. CAMFor User Manual v 3.35. National Carbon Accounting System. Australian Greenhouse Office, Technical Report No. 26, Canberra, Australia, p. 47. Richards, G.P., 2001. The FullCAM carbon accounting model: development, calibration and implementation for the National Carbon Accounting System. Australian Greenhouse Office, Australia. Richards, R., 1983. Should selection for yield in saline regions be made on saline or non-saline soils? Euphytica 32, 431-438. Rietz, D., Haynes, R., 2003. Effects of irrigation-induced salinity and sodicity on soil microbial activity. Soil Biology and Biochemistry 35, 845-854. Saha, S., Kudrat, M., Bhan, S., 1990. Digital processing of Landsat TM data for wasteland mapping in parts of Aligarh District (Uttar Pradesh), India. International Journal of Remote Sensing 11, 485-492.

38

Sardinha, M., Müller, T., Schmeisky, H., Joergensen, R.G., 2003. Microbial performance in soils along a salinity gradient under acidic conditions. Applied Soil Ecology 23, 237-244. Sarig, S., Steinberger, Y., 1994. Microbial biomass response to seasonal fluctuation in soil salinity under the canopy of desert halophytes. Soil Biology and Biochemistry 26, 1405-1408. Shaw, R., 1999. Soil salinity, electrical conductivity and chloride. Soil analysis: An interpretation manual. CSIRO Publishing, Melbourne, Victoria, Australia, 129–145. Shirato, Y., Hakamta, T., Taniyama, I., 2004. Modified Rothamsted carbon model for Andosols and its validation: Changing humus decomposition rate constant with pyrophosphate-extractable Al. Soil Science and Plant Nutrition 50, 149-158. Shirato, Y., Yokozawa, M., 2005. Applying the Rothamsted Carbon Model for long-term experiments on Japanese paddy soils and modifying it by simple tuning of the decomposition rate. Soil Science and Plant Nutrition 51, 405-415. Shirato, Y., Yokozawa, M., 2006. Acid hydrolysis to partition plant material into decomposable and resistant fractions for use in the Rothamsted carbon model. Soil Biology and Biochemistry 38, 812-816. Siderius, W., 1991. The use of remote sensing for irrigation management with emphasis on IIMI research concerning salinity, waterlogging and cropping patterns. Mission Report. International Institute for Aerospace Survey and Earth Sciences, The Netherlands, p. 97. Skene, T., Oades, J.M., 1995. The effects of sodium adsorption ratio and electrolyte concentration on water quality: laboratory studies. Soil Science 159, 65. Skjemstad, J., Janik, L., Taylor, J., 1998. Non-living soil organic matter: what do we know about it? Australian Journal of Experimental Agriculture 38, 667-680. Skjemstad, J., Spouncer, L., Cowie, B., Swift, R., 2004. Calibration of the Rothamsted organic carbon turnover model (RothC ver. 26.3), using measurable soil organic carbon pools. Australian Journal of Soil Research 42, 79-88. Smith, J., Gottschalk, P., Bellarby, J., Chapman, S., Lilly, A., Towers, W., Bell, J., Coleman, K., Nayak, D., Richards, M., Hillier, J., Flynn, H.C., Wattenbach, M., Aitkenhead, M., Yeluripurti, J.B., Farmer, J., Milne, R., Thomson, A., Evans, C., Whitmore, A.P., Falloon, P., Smith, P., 2010. Estimating changes in Scottish soil carbon stocks using ECOSSE. I. Model description and uncertainties. Climate Research 45, 179-192.

39

Smith, J., Smith, P., Monaghan, R., MacDonald, A., 2002. When is a measured soil organic matter fraction equivalent to a model pool? European Journal of Soil Science 53, 405-416. Smith, J., Smith, P., Wattenbach, M., Gottschalk, P., Romanenkov, V., Shevtsova, L., Sirotenko, O., Rukhovich, D., Koroleva, P., Romanenko, I., Lisovoi, N., 2007. Projected changes in the organic carbon stocks of cropland mineral soils of European Russia and the Ukraine, 1990–2070. Global Change Biology 13, 342-356. Smith, J., Smith, P., Wattenbach, M., Zaehle, S., Hiederer, R., Jones, R., Montanarella, L., Rounsevell, M., Reginster, I., Ewert, F., 2005. Projected changes in mineral soil carbon of European croplands and grasslands, 1990–2080. Global Change Biology 11, 2141-2152. Smith, P., Smith, J., Powlson, D., McGill, W., Arah, J., Chertov, O., Coleman, K., Franko, U., Frolking, S., Jenkinson, D., 1997. A comparison of the performance of nine soil organic matter models using datasets from seven long-term experiments. Geoderma 81, 153-225. Smith, P., Smith, J., Wattenbach, M., Meyer, J., Lindner, M., Zaehle, S., Hiederer, R., Jones, R., Montanarelia, L., Rounsevell, M., 2006. Projected changes in mineral soil carbon of European forests, 1990-2100. Canadian Journal of Soil Science 86, 159-169. Soil Survey Staff, 1998. Keys to soil taxonomy 8th ed. U.S. Govt Print. Office Washington, DC. Steven, M., Malthus, T., Jaggard, F., Andrieu, B., 1992. Monitoring responses of vegetation to stress. In: Cracknell, A., Vaughan, R. (Eds.), Remote sensing from research to operation: proceedings of the 18th annual conference of the Remote Sensing Society Nottingham, U.K., pp. 369-377. Sumner, M., Rengasamy, P., Naidu, R., 1998. Sodic soil: a reappraisal. Sodic Soils: Distribution, Properties, Management, and Environmental Consequences. Oxford University Press, Oxford, New York, USA, 3-17. Szabolcs, I., 1993. Soils and salinisation. In: Pessarakli, M. (Ed.), Handbook of Plant and Crop Stress. Marcel Dekker, New York, pp. 3–11 Tanji, K.K., 1990. The nature and extent of agricultural salinity problems. In: Tanji, K.K. (Ed.), Agricultural Salinity Assessment and Management Engineering Practice No. 71. American Society of Civil Engineers, New York, pp. 1–17. Tester, M., Davenport, R., 2003. Na+ tolerance and Na+ transport in higher plants. Annals of Botany 91, 503-527.

40

Theng, B., Tore, K., Sollins, P., 1989. Constituents of organic matter in temperate and tropical soils. In: Coleman, D., Oades, J., Uehara, G. (Eds.), Dynamics of Soil Organic Matter in Tropical Ecosystems. . NiffAL Project, University of Hawaii, pp. 5-32. Tisdall, J., Oades, J., 1982. Organic matter and water-stable aggregates in soils. European Journal of Soil Science 33, 141-163. Tripathi, S., Kumari, S., Chakraborty, A., Gupta, A., Chakrabarti, K., Bandyapadhyay, B.K., 2006. Microbial biomass and its activities in salt-affected coastal soils. Biology and Fertility of Soils 42, 273-277. Trumbore, S., Vogel, J., Southon, J., 1989. AMS 14C measurements of fractionated soil organic matter: an approach to deciphering the soil carbon cycle. Radiocarbon 31, 654–664. Turchin, P., 1998. Quantitative analysis of movement: measuring and modeling population redistribution in animals and plants. Sinauer Associates, Sunderland, USA. US Salinity Laboratory Staff, 1954. Diagnosis and improvement of saline and alkali soils. USDA Handbook No. 60. U.S. Government Printing Office Washington, DC. Venkataratnam, L., 1983. Monitoring of soil salinity in the Indo-Gangetic plain of NW India using multidate Landsat data. Proceedings of the 17th international symposium on remote sensing of the environment, Environmental Research Institute of Michigan, Michigan, USA, , pp. 369-377. Verma, K., Saxena, R., Barthwal, A., Deshmukh, S., 1994. Remote sensing technique for mapping salt affected soils. International Journal of Remote Sensing 15, 1901-1914. Vidal, A., Maure, P., Durand, H., Strosser, P., 1996. Remote sensing applied to irrigation system management: Example of Pakistan. EURISY colloquium: satellite observation for sustainable development in the Mediterranean area. Paris: promotion of education and information activities for the advancement of space technology and its application in Europe, pp. 132-142. Vincent, B., Vidal, A., Tabbet, D., Baqri, A., Kuper, M., 1996. Use of satellite remote sensing for the assessment of water logging or salinity as an indication of the performance of drained systems. In: Vincent, B. (Ed.), Evaluation of performance of subsurface drainage systems: 16th congress on irrigation and drainage, Cairo, pp. 203-216. Walpola, B., Arunakumara, K., 2010. Effect of salt stress on decomposition of organic matter and nitrogen mineralization in animal manure amended soils. Journal of Agricultural Sciences 5, 9-18. Wichern, J., Wichern, F., Joergensen, R.G., 2006. Impact of salinity on soil microbial communities and the decomposition of maize in acidic soils. Geoderma 137, 100-108.

41

Wong, V.N.L., Dalal, R.C., Greene, R.S.B., 2008. Salinity and sodicity effects on respiration and microbial biomass of soil. Biology and Fertility of Soils 44, 943-953. Wong, V.N.L., Dalal, R.C., Greene, R.S.B., 2009. Carbon dynamics of sodic and saline soils following gypsum and organic material additions: A laboratory incubation. Applied Soil Ecology 41, 29-40. Yuan, B.C., Li, Z.Z., Liu, H., Gao, M., Zhang, Y.Y., 2007. Microbial biomass and activity in salt affected soils under arid conditions. Applied Soil Ecology 35, 319-328. Zimmermann, M., Leifeld, J., Schmidt, M., Smith, P., Fuhrer, J., 2006. Measured soil organic matter fractions can be related to pools in the RothC model. European Journal of Soil Science 58, 658-667.

42

CHAPTER 2

Severity of salinity accurately detected and classified on a paddock scale with high resolution multispectral satellite imagery Raj Setia1, Megan Lewis2, Petra Marschner1, Ramesh Raja Segaran2, David Summers2,3 and David Chittleborough2

1Soils, School of Agriculture, Food and Wine, The University of Adelaide, Adelaide SA 5005, Australia

2School of Earth and Environmental Sciences, The University of Adelaide, Adelaide SA 5005, Australia

3Natural Resource and Economic Decision Sciences, CSIRO Ecosystem Sciences, PMB 2, Glen Osmond, SA 5064, Australia

The work contained in this chapter is published in Land Degradation & Development (With permission from Wiley-Blackwell, UK)

Setia, R., Lewis, M., Marschner, P., Raja Segaran, R., Summers, D., Chittleborough, D. (2011). Severity of salinity accurately detected and classified on a paddock scale with high resolution multispectral satellite imagery. Land Degradation & Development, DOI: 10.1002/ldr.1134

NOTE: Statements of authorship appear in the print copy of the thesis held in the University of Adelaide Library.

A Setia, R., Lewis, M., Marschner, P., Raja Segaran, R., Summers, D. & Chittleborough, D. (2011). Severity of salinity accurately detected and classified on a paddock scale with high resolution multispectral satellite imagery. Land Degradation & Development, Early View Online Version, Wiley Online Library

A NOTE:

This publication is included on pages 44-53 in the print copy of the thesis held in the University of Adelaide Library.

A It is also available online to authorised users at:

A http://dx.doi.org/10.1002/ldr.1134

A

54

CHAPTER 3

Is CO2 evolution in saline soils affected by an osmotic effect and calcium carbonate?

Raj Setia1*, Petra Marschner1, Jeff Baldock1,2 and David Chittleborough3

1Soils, School of Agriculture, Food and Wine, The University of Adelaide, Adelaide SA5005, Australia 2CSIRO Land and Water, Glen Osmond SA 5064, Australia

3School of Earth and Environmental Sciences, The University of Adelaide, Adelaide SA5005, Australia

The work contained in this chapter is published in Biology and Fertility of Soils (With permission from Springer)

Setia, R., Marschner, P., Baldock, J., Chittleborough, D., 2010. Is CO2 evolution in saline soils affected by an osmotic effect and calcium carbonate? Biology and Fertility of Soils 46, 781-792.

NOTE: Statements of authorship appear in the print copy of the thesis held in the University of Adelaide Library.

A Setia, R., Marschner, P., Baldock, J. & Chittleborough, D. (2010). Is CO2 evolution in saline soils affected by an osmotic effect and calcium carbonate? Biology and Fertility of Soils, v. 46 (8), pp. 781-792

A NOTE:

This publication is included on pages 56-67 in the print copy of the thesis held in the University of Adelaide Library.

A It is also available online to authorised users at:

A http://dx.doi.org/10.1007/s00374-010-0479-3

A

68

CHAPTER 4

Salinity effects on carbon mineralization in soils of varying texture

Raj Setia1*, Petra Marschner1, Jeff Baldock1, 2, David Chittleborough3, Pete Smith4 and Jo Smith4

1Soils, School of Agriculture, Food and Wine, The University of Adelaide, Adelaide SA5005, Australia 2 CSIRO Land and Water, Glen Osmond SA 5064, Australia 3School of Earth and Environmental Sciences, The University of Adelaide, Adelaide SA5005, Australia

4Institute of Biological and Environmental Sciences, School of Biological Sciences, University of Aberdeen, Aberdeen- AB24 3UU, Scotland, UK The work contained in this chapter is published in Soil Biology & Biochemistry

(With permission from Elsevier) Setia, R., Marschner, P., Baldock, J., Chittleborough, D., Smith, P., Smith, J. (2011). Salinity effects on carbon mineralization in soils of varying texture. Soil Biology & Biochemistry 43, 1908-1916

NOTE: Statements of authorship appear in the print copy of the thesis held in the University of Adelaide Library.

A Setia, R., Marschner, P., Baldock, J., Chittleborough, D., Smith, P. & Smith, J. (2011). Salinity effects on carbon mineralization in soils of varying texture. Soil Biology & Biochemistry, v. 43 (9), pp. 1908-1916

A NOTE:

This publication is included on pages 70-78 in the print copy of the thesis held in the University of Adelaide Library.

A It is also available online to authorised users at:

A http://dx.doi.org/10.1016/j.soilbio.2011.05.013

A

79

CHAPTER 5

Relationships between carbon dioxide emission and soil properties in salt-affected landscapes Raj Setia1*, Petra Marschner1, Jeff Baldock1,2 , David Chittleborough3 and Vipan Verma4

1Soils, School of Agriculture, Food and Wine, The University of Adelaide, Adelaide SA5005, Australia 2 CSIRO Land and Water, Glen Osmond SA 5064, Australia 3School of Earth and Environmental Sciences, The University of Adelaide, Adelaide SA5005, Australia

4Punjab Remote Sensing Centre, Ludhiana-141 004, Punjab, India

The work contained in this chapter is published in Soil Biology & Biochemistry (With permission from Elsevier)

Setia, R., Marschner, P., Baldock, J., Chittleborough, D., Verma, V., 2011. Relationships between carbon dioxide emission and soil properties in salt-affected landscapes. Soil Biology & Biochemistry 43, 667-674.

NOTE: Statements of authorship appear in the print copy of the thesis held in the University of Adelaide Library.

A Setia, R., Marschner, P., Baldock, J., Chittleborough, D. & Verma, V. (2011). Relationships between carbon dioxide emission and soil properties in salt-affected landscapes. Soil Biology & Biochemistry, v. 43(3), pp. 667-674

A NOTE:

This publication is included on pages 81-88 in the print copy of the thesis held in the University of Adelaide Library.

A It is also available online to authorised users at:

A http://dx.doi.org/10.1016/j.soilbio.2010.12.004

A

89

CHAPTER 6

Introducing a decomposition rate modifier in the Rothamsted carbon model to predict soil organic carbon stocks in saline soils Raj Setia1, Pete Smith2, Petra Marschner1*, Jeff Baldock1,3, David Chittleborough4 and Jo Smith2

1Soils, School of Agriculture, Food and Wine, The University of Adelaide, Adelaide SA5005, Australia 2 Institute of Biological and Environmental Sciences, 23 St Machar Drive, University of Aberdeen, Aberdeen- AB24 3UU, Scotland, UK 3 CSIRO Land and Water, Glen Osmond SA 5064, Australia 4School of Earth and Environmental Sciences, The University of Adelaide, Adelaide SA5005, Australia The work contained in this chapter is published in Environmental Science & Technology

(With permission from American Chemical Society) Setia, R., Smith, P., Marschner, P., Baldock, J., Chittleborough, D., Smith, J. (2011). Introducing a decomposition rate modifier in the Rothamsted carbon model to predict soil organic carbon stocks in saline soils. Environmental Science and Technology, dx.doi.org/10.1021/es200515d

NOTE: Statements of authorship appear in the print copy of the thesis held in the University of Adelaide Library.

Setia, R., Smith, P., Marschner, P., Baldock, J., Chittleborough, D. and Smith, J. (2011) Introducing a decomposition rate modifier in the Rothamsted carbon model to

predict soil organic carbon stocks in saline soils

Environmental Science and Technology, v.45 (15), pp. 6396–6403, August 2011

NOTE: This publication is included in the print copy of the thesis

held in the University of Adelaide Library.

It is also available online to authorised users at:

http://dx.doi.org//10.1021/es200515d

99

CHAPTER 7

Simulation of salinity effects on soil organic carbon: past, present and future carbon stocks Raj Setia1, Pete Smith2, Petra Marschner1*, Pia Gottschalk3, Jeff Baldock1,4, Vipan Verma5 and Jo Smith2

1Soils, School of Agriculture, Food and Wine, The University of Adelaide, Adelaide SA5005, Australia 2 Institute of Biological and Environmental Sciences, 23 St Machar Drive, University of Aberdeen, Aberdeen- AB24 3UU, Scotland, UK 3 Potsdam Institute for Climate Impact Research, Telegrafenberg, P.O. Box 601203, 14412 Potsdam, Germany 4 CSIRO Land and Water, Glen Osmond SA 5064, Australia 5Punjab Remote Sensing Centre, Ludhiana-141 004, Punjab, India The work contained in this chapter has been submitted to Agriculture, Ecosystems & Environment.

NOTE: Statements of authorship appear in the print copy of the thesis held in the University of Adelaide Library.

102

A Setia, R., Smith, P., Marschner, P., Gottschalk, P., Baldock, J., Verma, V. & Smith, J. (2012) Simulation of salinity effects on past, present and future soil organic carbon stocks Environmental Science & Technology, v. 46 (3), pp. 1624-1631

A NOTE:

This publication is included on pages 102-127 in the print copy of the thesis held in the University of Adelaide Library.

A It is also available online to authorised users at:

A http://dx.doi.org/10.1021/es2027345

A

A NOTE

Published article is titled: 'Simulation of salinity effects on past, present and future soil organic carbon stocks'.

128

CHAPTER 8

CONCLUSIONS AND FUTURE RESEARCH

The global carbon (C) pool up to 1 m depth, comprising organic and inorganic C is 2300

Pg, of which 67.4% (1550 Pg) is organic and 32.6% (750 Pg) is inorganic (Lal, 2008).

Today, the soil organic carbon (SOC) content is lower than before human intervention

due to land use changes and the resulting carbon dioxide (CO2) emissions have played

an important role in enrichment of atmospheric carbon dioxide (Smith et al., 2008).

This is particularly the case for degraded soils, e.g. salt-affected soils. Hence, these

soils have a significant potential for increasing C sequestration in the above-ground

biomass and in the SOC pool through amelioration (Lal, 2001).

The total area of saline and sodic soils is 397 and 434 million ha, respectively

which is approximately 6.5 % of the world land area (Szabolcs, 1993). Adoption of soil

restorative measures in these soils requires the integrated use of frontier technologies

such as remote sensing, experiments and modelling which can be used to estimate the

potential of these soils for SOC sequestration.

Salinization may be quite localized and change from one season to another.

Therefore, it is necessary to map the spatial and temporal changes in its occurrence.

Such mapping can be accomplished in several ways. Since field and laboratory

methods are quite expensive and time-consuming, cheaper and less time-consuming

methods such as remote sensing can play an important role in identifying, mapping

and monitoring of soil salinity (Kalra and Joshi, 1996, Metternicht and Zinck, 2003;

Shrestha, 2006; Dutkiewicz et al., 2009).

The unsupervised classification of whole image of the Australian study area was

unable to discriminate between severity of salinity because the expression of salinity in

this area differs from paddock-to-paddock due to crop and management factors. The

paddock-by-paddock-approach takes into account this variability and could

differentiate three classes of salinity which were significant related with electrical

conductivity (EC) but not with sodium adsorption ratio (SAR). This clearly shows that in

salt-affected areas with small localised patches of salinity and differential land

management, the paddock-by-paddock approach can be used for accurate mapping.

129

With respect to the effect of salinity on decomposition rates, the studies

reported in this thesis highlighted two important points. Firstly, that incubation studies

where salt is added may overestimate the effect of salinity on microbial activity and

thus decomposition rates. And secondly, that the EC may not be a good measure for

the salinity stress experienced by microbes in saline soils. Addition of salt to soils will

cause an immediate osmotic stress to soil microbes which may not give them sufficient

time to adjust physiologically or via changes in community composition to the lower

osmotic potential. In the field, salinity develops more slowly and therefore allows for

adjustment. Therefore, to evaluate the effect of salinity on decomposition rates, saline

soils from the field should be used. The EC measured in fixed soil-to-solution ratio is a

poor indicator of the salt stress experienced by plants and microbes because as soils

dry, the salt concentration in the soil solution increases (decreasing osmotic potential).

Thus, the osmotic potential varies with water content in a given soil but also between

soils. Two soils of different texture may have the same EC1:5 (EC of 1:5 soil: water

suspension), but the osmotic potential is lower in the soil with lower water content.

The relative importance of soil properties may be quite different in salt-affected soils

to that in non-salt-affected soils. The regression analysis of the CO2 emission from non-

salt-affected soils and salt-affected soils collected from India and Australia showed that

cumulative CO2-C emission was significantly positively correlated with particulate

organic carbon (POC) and humus-C but significantly negatively related with EC and

significantly positively with osmotic potential. Therefore, salinity strongly affects

decomposition rates and should be taken account into SOC models for studying SOC

dynamics in salt-affected soils.

When modelling SOC turnover in soils, it is important to not only consider

decomposition rates but also plant inputs. There are many studies showing reduced

yield in saline soils, therefore, plant inputs into saline soils will be reduced. The

Rothamsted Carbon Model (RothC, Jenkinson et al. 1987, Coleman and Jenkinson

1996) is often used for estimations of future SOC stocks and CO2 emissions, but it had

not been calibrated for saline soils.

In the work described here, reduced plant inputs were calculated based on

data from Maas and Hoffman (1977) who compiled a list of threshold and slope values

for yield of various crops. The importance of considering reduced plant inputs into

130

saline soils was demonstrated by modelling past SOC stocks considering reduced

decomposition only or reduced decomposition and reduced plant inputs. Using the

reduced decomposition only, the past SOC content was lower than the present,

suggesting an increase in SOC due to salinity. On the other hand, including both

reduced decomposition and reduced plant inputs showed that saline soils had lost

considerable amounts of SOC since they became saline. The average loss of SOC due

to salinity was 55 t ha-1 for Australian soils and 31 t ha-1 for Indian soils. Since this and

other study showed that saline soils have lower SOC contents compared to non-saline

soils, only the latter simulation (reduced decomposition and reduced plant inputs) is

correct. The iterative approach to calculate past SOC stocks is novel and useful.

However its limitation is that it cannot be verified and it is not known when the now

saline soils were non-saline.

The modelling results with the decomposition rate modifier and plant inputs

modifier indicated a greater decrease in SOC in saline than in non-saline soils of India

and Australia in the future climate. Thus, the previous projections of the decrease of

SOC in saline soils have been underestimated; this also means that saline soils will emit

more CO2 than previously thought.

This thesis represents development of the following innovative and unique

approaches:

1. A paddock-by-paddock approach for accurate mapping of salinity for

smaller area using multispectral high resolution imagery.

2. Using osmotic potential (not EC) for estimating the effect of salinity on

microbial activity and soil respiration.

3. Calculation of a decomposition rate modifier for saline soils and its inclusion

in Roth C

4. Development of an approach to calculate plant inputs into saline soils and

its inclusion in Roth C

5. A novel approach for calculating past SOC stocks when saline soils were

non-saline.

131

Recommendations for future research work

1. The rate modifier for decomposition was derived from CO2 emission over 120

days in a single incubation experiment with 120 soils. The decomposition rate

modifier derived from this limited set of soils may not be generally applicable.

Additionally, there was no relationship between sodicity and CO2 emission in

this study, which may be due to the limited range of sodicity in the sampled

soils. Therefore, a similar incubation experiment should be conducted focussing

specifically on sodic soils. In order to preserve the soil structure, such an

experiment would have to be carried out with intact soil cores.

2. Due to the short incubation time (120 days), only easily decomposable (labile)

SOC pools changed. It was assumed that the same rate modifier applies for all

SOC pools. This issue could be addressed by analysing archived soil samples

from salt-affected areas taken at least 10 years previously and comparing the

SOC pools of these soils with those of present soils from the same location.

3. The plant input modifier used in modelling was based on a generalised

equation between EC and yield and did not take into account the differential

sensitivity of plants to salinity. Moreover, yield may not be directly reflected in

C inputs, particularly with respect to root-derived C. Root and shoot biomass of

common crop plants will need to be determined over a range of salinity and

sodicity levels to develop a more accurate equation for plant inputs into saline

soils.

4. To verify if the decomposition rate modifier is applicable to the field situation,

CO2 emission from salt-affected soils should be measured in the field,

accompanied by soil sampling to determine EC and water content.

References

Coleman K, Jenkinson D, 1996. RothC-26.3-A Model for the turnover of carbon in soil. RothC-26.3-A Model for the turnover of carbon in soil. In: Powlson, D., Smith, P., Smith, J. (Eds.), Evaluation of soil organic matter models using existing long-term datasets. NATO ASI SERIES, 38, 237-246.

132

Dutkiewicz A, Lewis M, Ostendorf B, 2009. Evaluation and comparison of hyperspectral imagery for mapping surface symptoms of dryland salinity. International Journal of Remote Sensing 30: 693-719.

Jenkinson D, Hart P, Rayner J, Parry L, 1987. Modelling the turnover of organic matter in long-term experiments at Rothamsted. Intecol. Bull 15: 1-8.

Kalra N, Joshi D, 1996. Potentiality of Landsat, SPOT and IRS satellite imagery, for recognition of salt affected soils in Indian Arid Zone. International Journal of Remote Sensing 17: 3001-3014.

Lal R, 2001. Potential of desertification control to sequester carbon and mitigate the greenhouse effect. Climatic Change 51: 35-72.

Lal R, 2008. Carbon sequestration. Philosophical Transactions of the Royal Society B: Biological Sciences 363: 815-830. Mass E, Hoffman G, 1977. Crop salt tolerance current assessment. Journal of the Irrigation and Drainage Division ASCE 504: 115-135.

Metternicht G, Zinck J, 2003. Remote sensing of soil salinity: potentials and constraints. Remote Sensing of Environment 85: 1-20.

Shrestha R, 2006. Relating soil electrical conductivity to remote sensing and other soil properties for assessing soil salinity in northeast Thailand. Land Degradation and Development 17: 677-689.

Smith P, Fang C, Dawson JJC, Moncrieff JB, 2008. Impact of global warming on soil organic carbon. Advances in Agronomy 97: 1-43.

Szabolcs I, 1993. Soils and salinisation. In: Pessarakli M, (ed.) Handbook of Plant and Crop Stress. Marcel Dekker, New York, 3–11.

133

APPENDIX -I

Mapping of salt affected soils in Muktsar district, Punjab, India

The multi-temporal geo rectified Resourcesat-I LISS-III data (spatial resolution = 23.5

m) acquired during spring (April 2005), autumn (October 2005) and winter (February

2006) of 2005-06 was used for mapping of salt-affected soils in Muktsar district. The

imagery was provided as four bands [green (0.52 - 0.59 µm), red (0.62 - 0.68 µm), near-

infrared (0.76 – 0.86 µm) and short-wave infrared (1.55 – 1.70 µm)]. Ancillary data in

the form of Survey of India (SOI) topographical maps, existing waste land data and

other published reports were used as reference data. Survey of India topographical

maps on 1:50,000 scale were used for identification of base features and for planning

of the ground truthing. Ground-truthing was performed by collecting 120 soil samples

over the study area of 2631 km2.

The methodology is based on on-screen interpretation using standard image

interpretation keys like tone, texture, size, pattern, association etc. which was

followed by ground truthing. To delineate indicators of salinity like reduced plant cover

and growth, halophytic vegetation, greater soil exposure and possibly brighter soil

surfaces, suitable band combinations were used where the signatures of particular

class were quite evident. The standard false colour composites (FCC) showed white

patches of varying tones in the red background indicating the salt efflorescence of salt

affected soils which showed higher digital values than non-salt-affected soils. Similar

tone was also observed for sandy soils prevalent in the area but these soils were

discriminated by associated features and ground truthing. The salt-affected soils were

mapped in the light of the ground observations, visual interpretation keys and

available soil parameters (electrical conductivity and sodium adsorption ratio) (Figure

1). The salt-affected soils occupy less than 1% area of district in various seasons and

the salinity was mainly induced by irrigation (secondary salinization). Some the patches

were observed along the twin canals, these canals are aligned across the natural

gradient of the area, obstructing the natural flow of water. Nearly 61% area of the

district has poor quality groundwater, with 37% of the area having saline to highly

saline, 19% saline-sodic and 5% area sodic waters (Sharma et al., 2003).

134

Figure 1: Classified Resourcesat-I LISS-III data of the Muktsar district, Punjab, India.

References

Sharma, P.K., Verma, V.K., Litotia, P.K., Sood, Anil., Loshali, D.C., Kumar, Ashok., Singh Charanjit., Bhatt, C.M. and Chopra, Rajiv. (2003). Resource Atlas Muktsar District, Punjab Remote Sensing Centre Publication, Ludhiana, India.

NOTE: This figure is included on page 134 of the print copy of the thesis held in the University of Adelaide Library.